JP3486185B1 - Performance information compression device, performance information decoding device, performance information compression method, performance information decoding method, performance information compression program, performance information decoding program - Google Patents

Abstract

Abstract: [PROBLEMS] To recover the order of each piece of performance information even after decoding, to prevent a situation in which the decoded performance information does not match the original performance information, and to record a part of the performance information. Reduce the size of the primary code by omitting it. A performance information compression device or the like according to the present invention uses an SMF or the like for each channel for at least information on a relative time between events, information on a sounding start pitch, information on a sound intensity, and information on a sounding end pitch. A primary code generation unit 2 that generates a primary code in which each of the information is separated into independent information regardless of the channel, and the collected information is arranged in an independent area. And a secondary code generation unit 3 for compressing information of each area of the primary code generated by the primary code generation unit 2 by a decoding method such as an LZ encoding method and a Huffman encoding method.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technique for compressing or decoding performance information and the like, and particularly to a performance information compression device, a performance information decoding device, and a performance information compression device for efficiently compressing the amount of performance information data. The present invention relates to a method, a performance information decoding method, a performance information compression program, and a performance information decoding program.

[0002]

2. Description of the Related Art Conventionally, MIDI (Musical Instrument Digital Interface) has been used as a standard relating to an interface for digitally communicating performance information from a musical instrument or the like.
ce) standards are generally known. And this MIDI
As a file to save the data defined by the standard,
For example, a standard MIDI file (hereinafter referred to as SMF) is generally known and is already widely used in various fields.

In this SMF, each performance information is represented by a MIDI message in which several 1-byte characters are combined, and the MIDI message is
The MIDI standard defines a 1-byte status,
This distinguishes the performance information. Further, the status byte is followed by data bytes, and the number of each data byte differs depending on the type of performance information. In addition to the above, the status byte contains information related to the status type and channel number.

However, in view of the fact that the SMF file size becomes relatively large today, when the SMF is applied to, for example, a system for recording and transmitting a large number of music data, the SMF is applied to the SMF. It is hoped that the amount of data of each performance information included will be efficiently compressed, and technical development for that purpose is underway.

As a reversible coding method which is a coding method for decoding the coded information and completely matching the original information, for example, the LZ (Lempel-Zif) coding method and the Huffman code are used. A coding method, a run length coding method, and a method in which these are combined (for example, LHA, ZIP, etc.) are generally known and widely used.

Of these, first, the LZ method compresses the repeated portion by utilizing the repetition of the data pattern, and therefore, the same data pattern appearing in a predetermined range in the input data. When there are many numbers, the compression rate becomes low. Here, the compression rate means the ratio of the data size after compression to the data size before compression. And, the reduction of the compression rate means that the compression efficiency is increased (improved).

Further, the Huffman coding system constructs a predetermined tree that is weighted by the frequency of appearance of the code for each byte regardless of the data pattern so that a more weighted code can be recorded with a smaller number of bits. Since it is encoded into, it is effective for data in which the appearance frequency of codes for each byte is biased.

The LHA coding method which is a combination of these LZ method and Huffman coding method has all of the advantages of the above LZ method and Huffman coding method.

Here, in the SMF, each performance information is composed of a delta time (hereinafter referred to as Δtime) and an event, and each event has a note-on or note-off status, pitch, and note. It includes information such as the strength of.

In general, music is often composed of repetitions of similar phrases, but the music producer slightly changes, for example, the sound intensity or the sound length in order to avoid the music becoming monotonous. There are cases. Also, even if phrases with the same pattern exist in different channels, SM
If the event of each performance information of F includes the information related to the channel number, the recorded data will have a different pattern.

As described above, in the SMF, in order to improve the compression efficiency in the case of compressing by the general-purpose compression method from the above-mentioned circumstances caused by the fact that the event, which is the minimum unit of data recording, includes a plurality of information. Further ingenuity is needed.

In view of this point, for example, Japanese Patent Laid-Open No. 9-16
According to Japanese Patent No. 168, performance information is separated into pitch information, sound intensity information, sound length information, and other information by a primary code generating means, and each information is arranged in an independent area. By doing so, a technique relating to a performance information compression device and a performance information decoding device, which is characterized by increasing the compression efficiency by the LZ method thereafter, is disclosed.

[0013]

However, in the technique disclosed in the above-mentioned Japanese Patent Laid-Open No. 9-16168, the performance information is made independent in spite of being temporally independent at the time of primary encoding. Since the information related to the order of the performance information is not stored, the order of the performance information may be changed when the encoded performance information is decoded.
That is, there may be a case where the decoded performance information does not completely match the original performance information.

The present invention has been made in view of the above problems, and an object thereof is to recover the order of performance information even after decoding so that the decoded performance information does not match the original performance information. To prevent situations such as
It is possible to reduce the size of the primary code by omitting recording of some performance information, and further to efficiently compress or decompress the data amount of performance information in combination with a general-purpose compression method. A performance information compression device, a performance information decoding device, a performance information compression method, a performance information decoding method, a performance information compression program, and a performance information decoding program.

[0015]

In order to achieve the above-mentioned object, the invention of claim 1 is an event consisting of at least sounding start pitch information, sound intensity information, sounding end pitch information, and other information. And a file having a plurality of track data including information of relative time between the events, and a performance information compression device for compressing the file, the plurality of track data of the file are
Separating the belt rack every data attributable to each channel, decoding
The information of the track data belonging to each channel is retained while retaining the information indicating the attribution to the channel for reference when converting.
The information is further separated into each type, and the information for each of the separated types
A summary for each different species, a primary code generating means for generating a primary code by placing in a separate area for each separate species, arranged in each region of the primary code generated by the primary code generation means And a secondary code generation means for compressing each of the generated information by a predetermined compression method. The invention of claim 2 is at least a sound
Information, sound intensity information, sound duration information, etc.
Information consisting of information and relative time information between the events
A file with multiple track data containing
Performance information compression that receives the input of the file and compresses the file
In the device, multiple track data of the above file
Is separated for each track data belonging to each channel,
Information indicating attribution to the channel for reference when decoding
Track data belonging to each channel while maintaining information
Information is further separated for each type, and for each separated type
Information is grouped by type and placed in an independent area for each type.
A primary code generating means for generating a primary code by
In each area of the primary code generated by the primary code generation means
Secondary that compresses each placed information by a predetermined compression method
And a performance information pressure
This is a compression device. The invention of claim 3 is the above primary
The code generation means is based on the information on the relative time between the events.
Based on the above track data, multiple events related to the same time
If it is determined that the
Event recorded at the end of the event group consisting of
Or there is a single event for a given time
For the event is to determine the case, to the channel
It is further recommended to exclude from the subject that holds the information showing attribution.
The performance information compression apparatus according to claim 1 or 2,
It is what The invention of claim 4 is the above-mentioned primary code raw.
The composition means uses the event type identification code as the track data.
Assigned to each event included in
It is recommended to put identification codes together and place them in independent areas.
The performance information compression apparatus according to claim 1 or 2, wherein
It is a place. The invention of claim 5 is the above-mentioned primary code.
The means of generation is the event recorded in the performance information file.
Of the events included in the group
The primary code is the identification information of the event type of the high frequency event.
A further feature is that it is included in the relative time code of the
The performance information compression device according to claim 1 or 2.
Of. According to a sixth aspect of the present invention, the primary code generating means is provided.
If the above event is of a given type,
Event class information instead of information indicating belonging to Nell
The performance of claim 2, further characterized by recording
This is a performance information compression device. The invention of claim 7 is
Generated by the performance information compression device according to claim 1.
The generated secondary code into a primary code by a predetermined decoding method
Then, the performance information is restored by decoding the primary code into a file.
No. 2 device for decoding the above secondary code into a primary code 2
An independent area is provided for the next code decoding means and the primary code.
Extracts each type of information placed in the area for each type
Then, while referring to the information showing attribution to the above channel,
Each piece of information output is assigned to multiple track data belonging to each channel.
Data as a track data of the file.
Primary code decoding means for decoding into a file by configuring
And a performance information decoding device characterized by having
It is a thing. The invention of claim 8 is the performance according to claim 2.
Predetermine the secondary code compressed and generated by the information compression device
To the primary code by the decoding method of
In the performance information decoding device for decoding the
A secondary code decoding means for decoding the secondary code into a primary code;
Regarding the primary code, seeds placed in independent areas
Extract each type of information for each type,
While referring to the information indicating attribution, each extracted information is
It is configured as multiple track data belonging to the
File as track data of the file.
And a primary code decoding means for decoding into a file.
It is used as a performance information decoding device. Claim 9
According to the invention, the primary code decoding means is
Is the type of each event placed in an independent area
Is the event type identification code assigned to each event.
Identifying further based on the number.
The performance information decoding apparatus described in 7 or 8 is used.
According to a tenth aspect of the present invention, the primary code decoding means is an event
Channel is of a given type, the
Refer to event class information instead of the information indicating the genus
It is a further feature that the two track data are configured.
The performance information decoding apparatus according to claim 8 is used.
According to the invention of claim 11, at least information on the pitch at which the pronunciation is started,
Information about the strength of the sound, information about the ending pitch, or other information
And the information of the relative time between the events.
Import a file with multiple track data containing
The performance information compression method that receives the force and compresses the file
In the above, by the primary code generation means,
Track data of each track
Channel to be separated for each data and used for reference when decoding
Belonging to each channel while retaining the information indicating attribution to
The information of the track data that is
Information on each separated type is collected for each type and independent for each type
To generate a primary code by arranging in a specified area
And the secondary code generation means causes the primary code generation means to
Each information placed in each area of the primary code generated by
Compressing with a predetermined compression method.
The performance information compression method is characterized by Contract
The invention of claim 12 requires at least pitch information and sound intensity.
Event consisting of information, duration information, and other information
And a plurality of information including relative time information between the events.
Input the file containing the track data of
In the performance information compression method that compresses a file, the primary code
The No. generation means, the track data of several of the file
Data is separated for each track data belonging to each channel
And show attribution to the channel for reference when decoding.
Track information belonging to each channel
The data of the data is further separated for each type, and the separated type
Information for each type is collected for each type and placed in an independent area for each type.
Generating a primary code by placing the secondary code
Generated by the generating means by the primary code generating means
Compressed each information placed in each area of the primary code
Compressing by the method, and
This is a performance information compression method. From claim 13
Ming is between the events by the primary code generation means.
Same time as the above track data based on relative time information
When it is determined that multiple events related to are recorded
At the end of the event group consisting of the above multiple events
Events recorded, or events related to a predetermined time
If it is determined that there is only one event,
From the target that holds the information that indicates attribution to the channel
Further comprising the step of excluding.
The performance information compression method described in item 11 or 12 is used.
is there. According to a fourteenth aspect of the present invention, there is provided the above primary code generation means.
The event type identification code is included in the above track data.
Assigned to each event that is generated, the event type identification mark
The steps to assemble the issues and place them in independent areas.
13. The method according to claim 11 or 12, characterized in that
This is a performance information compression method. Invention of Claim 15
Is recorded in the performance information file by the primary code generating means.
Of the events included in the recorded event group,
Event type of at least the most frequently used event
The identification information of is recorded in the relative time code of the primary code.
The method further comprising the step of:
The performance information compression method described in 11 or 12 is used.
It According to a sixteenth aspect of the present invention, there is provided the above primary code generation means.
If the above event is of a given type,
Event class information instead of information indicating belonging to Nell
And a step of recording
The performance information compression method according to claim 12 is used.
The invention of claim 17 is the performance information compression according to claim 11.
Decoding the secondary code compressed and generated by the method
Decodes the primary code by law, further phi the primary code
In the performance information decoding method for
Means for decoding the secondary code into a primary code.
And the primary code decoding means
Information of each type placed in an independent area.
Information indicating the attribution to the above channels is extracted for each type.
While referring to each extracted information,
Number of track data, and the file
File that is decrypted into a file by configuring
And a performance information decoding method comprising:
It is what The invention of claim 18 relates to claim 12.
Secondary generated by compression using the performance information compression method described
The code is decoded into a primary code by a predetermined decoding method, and
In a performance information decoding method for decoding the primary code into a file
Then, the secondary code is decoded by the secondary code decoding means.
And the primary code decoding means.
Regarding the primary code, seeds placed in independent areas
Extract each type of information for each type,
While referring to the information indicating attribution, each extracted information is
It is configured as multiple track data belonging to the
File as track data of the file.
And decrypting into a file.
This is a performance information decoding method. Invention of Claim 19
When the decoding is performed by the primary code decoding means,
Is the type of each event placed in an independent area
Event type identification assigned to each event
The method further comprises the step of identifying based on the code.
The performance information decoding method according to claim 17 or 18,
It is what The invention according to claim 20 is the above primary code
If the event is of a given type by the decryption means
Instead of the information showing attribution to the above channel,
Class that composes track data while referring to class information
19. The method according to claim 18, further comprising a step.
This is the performance information decoding method. From claim 21
Akira means at least information about the starting pitch and information on the strength of the sound.
Event, consisting of information, end tone pitch information, and other information
And information about the relative time between the events.
The input of a file containing several track data
A performance information compression program that compresses the file
On your computer.
The chromatography data, separation for each track data belonging to each channel
And show attribution to the channel for reference when decoding.
Track information belonging to each channel
The data of the data is further separated for each type, and the separated type
Information for each type is collected for each type and placed in an independent area for each type.
The step of generating a primary code by
Each information placed in each area of the generated primary code is given a predetermined pressure.
A step of compressing by a shrinking method, and
This is a performance information compression program. Claim 22
Invention, at least pitch information, sound intensity information, sound
Event consisting of length information and other information
Multiple traps containing information on relative time between vents and
Input the file containing the data
The performance information compression program that compresses
Data of multiple tracks in the above file
Decode by separating each track data belonging to each channel
Information that indicates attribution to the channel for reference when
Information of track data belonging to each channel while being retained
The information is further separated into each type, and the information for each of the separated types
Can be grouped by type and placed in an independent area for each type.
And a step of generating a primary code with
Each information placed in each area of the next code is converted into a predetermined compression method.
Performance information for executing more compression steps and
It is a compression program. Invention of Claim 23
Of the relative time between the events to the computer
Based on the information, multiple items related to the same time in the above track data
If it is determined that the event of
Recorded at the end of an event group consisting of several events
There is only one event or event related to a given time
If it is determined that the event is
Exclude from holding the information indicating attribution to
22. The step is further executed.
Or the performance information compression program described in 22
is there. According to a twenty-fourth aspect of the present invention, the computer is provided with an event
The event type identification code is used for each event included in the track data.
Assigned to each event and put the event type identification code
And then perform the step of placing it in a separate area.
23. The performance according to claim 21 or 22, characterized in that
This is an information compression program. Claim 25
Akira is recorded in the performance information file on the computer.
Among the events to be included in the event group that has been, at least
The event type of the most frequently used event
To record information by including it in the relative time code of the primary code
3. The method according to claim 2, further comprising executing the steps.
The performance information compression program described in 1 or 22
Is. The invention of claim 26 is the above computer,
If the event is of a given type, go to the channel
Record event class information instead of attribution information
The method further comprising the step of:
It is the performance information compression program described in Item 22.
It The invention of claim 27 is the performance information according to claim 21.
The secondary code compressed and generated by the compression program
Decoding to a primary code by a fixed decoding method,
In the performance information decoding program that decodes the issue into a file
And the computer decodes the secondary code into a primary code.
And the primary code above
Extract each information for each type that is placed, for each type,
While referring to the information showing attribution to the above channel,
Multiple track data that attribute each information to each channel
As the track data of the file.
Perform the steps of decrypting to a file by
A performance information decoding program characterized by
Of. The invention of claim 28 is the operation of claim 22.
Secondary code compressed and generated by the performance information compression program
Signal into a primary code by a predetermined decoding method, and further
Performance information decoding program for decoding primary code into file
In the computer, the secondary code is converted into the primary code.
Independent of the decoding step and the above primary code
Extract each type of information placed in the area for each type.
While referring to the information showing attribution to the above channel,
Multiple tracks that attribute each extracted information to each channel
Configured as data, and also as file track data
And decrypting into a file by configuring
A performance information decoding program characterized by being executed;
It was done. The invention according to claim 29 is the above computer.
In the above decoding, it is placed in an independent area.
The type of each event is assigned to each event.
And the step of identifying based on the event type identification code
28. The method according to claim 27 , further comprising:
The performance information decoding program described in No. 8 is used.
According to the invention of claim 30, the computer is provided with an event.
If it is of a certain type, please attribute to the above channel.
Refer to the event class information instead of the information shown
Performing further steps to configure rack data
29. The performance information decoding program according to claim 28,
It is a mess.

[0016]

[0017]

[0018]

[0019]

[0020]

[0021]

[0022]

[0023]

[0024]

[0025]

[0026]

[0027]

[0028]

[0029]

[0030]

[0031]

[0032]

[0033]

[0034]

[0035]

[0036]

[0037]

[0038]

[0039]

[0040]

[0041]

[0042]

[0043]

[0044]

[0045]

[0046]

[0047]

[0048]

[0049]

[0050]

[0051]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings.

(First Embodiment) First, a performance information compression apparatus, a performance information decoding apparatus, a performance information compression method, a performance information decoding method, a performance information compression program, and a performance information decoding apparatus according to the first embodiment. The format of input data (music file) that is assumed in the program will be described in detail. In the first embodiment, the SMF format is assumed as the music file format. The SMF is roughly divided into a header chunk and a track chunk, and each header chunk contains minimum information directly related to the entire MIDI file, for example, information such as the number of track data. Furthermore, each track chunk can include MIDI data (track data) for up to 16 channels. Each track data is composed of two elements such as Δtime and event. More specifically, this Δtime is stored in variable length notation and indicates the time interval until the subsequent event. That is, if the first event of a track occurs at the same time as the start of the track, or if two events occur at the same timing, the Δtime becomes 0. Further, the event includes various performance information such as note-on or note-off status, pitch (note number), and sound intensity (velocity). That is, this event is a MIDI channel message, and more specifically,
It consists of a status byte and a data byte. If the status is the same as the preceding event, the status byte may be omitted. However, the first event must specify the status.

The above is an outline of the format of the SMF. In the format adopted by the performance information compression apparatus and the like according to the first embodiment, the events included in the event group recorded in the performance information file are included. Out of
A further feature is that at least the identification information of the event type of the most frequently used event is included in the relative time code of the primary code and recorded. More specifically, the Δ time of the primary code is converted and recorded so that the event type can be identified, so that the event type identification code can be encoded when the event with the highest frequency of use and the event of the next highest frequency of use are encoded. It is omitted to reduce the data amount of the primary code. This is, of course, one of the characteristic portions according to the first embodiment.

(Performance information compression device, etc.) Hereinafter, referring to FIG. 1 to FIG.
The configuration and operation of the performance information compression device, performance information compression method, and performance information compression program according to the first embodiment that generate the next code will be described in detail.

The operation of this performance information compression device is as follows.
It corresponds to the performance information compression method and the performance information compression program according to the first embodiment. The performance information compression program can be executed by a general-purpose computer or the like.

First, FIG. 1 shows the configuration of a performance information compression apparatus according to the first embodiment of the present invention, and will be described. Figure 1
As shown in FIG. 1, the performance information compression device 1 includes a primary code generation unit 2 and a secondary code generation unit 3. The primary code generating means described in the claims corresponds to the primary code generating section 2 and the like, and the secondary code generating means in the claims corresponds to the secondary code generating section 3 and the like. . However, it is not limited to this.

The primary code generator 2 rearranges the performance information contained in the music file 4 in the above-mentioned format to generate the primary code 5 in order to improve the compression efficiency of the secondary code generator 3. It is a thing. Then, the secondary code generation unit 3 compresses the primary code 5 by a general-purpose compression method,
The secondary code 6 is generated.

That is, the primary code generation unit 2 separates the track data of the file into a plurality of track data belonging to each channel, and holds the information indicating the belonging to the channel while belonging to each channel. The track data is further separated for each of the above information, the separated information is grouped for each type of information regardless of belonging to the channel, and is arranged in an independent area for each type. The code 5 is generated. The secondary code generation unit 3 uses the primary code 5
The secondary code 6 is generated by compressing each piece of information arranged in each area of 1 by a predetermined compression method.

In the first embodiment, as a general-purpose compression method adopted in the secondary code generator 3, for example, LZ is used.
Encoding method, Huffman encoding method, run-length encoding method,
Furthermore, a method that combines these (for example, LHA or ZIP
Etc.) is adopted, but the invention is not limited to these.

Here, the detailed configuration of the primary code generator 2 is as shown in FIG.

That is, the primary code generating section 2 includes a track separating section 7, a channel separating and channel number code generating section 8, a code generating section 9 other than the channel number, and a code arranging section 10.
have. Hereinafter, the configuration and operation of each of these units 7 to 10 will be described in detail.

First, the detailed structure of the track separating section 7 is as follows.
As shown in FIG.

That is, the track separation section 7 has a file analysis section 7a.

The file analysis unit 7a determines whether or not the track chunk of the music file 4 includes a plurality of track data, based on the information on the number of track data included in the header chunk of the music file 4. To do. When it is determined that a plurality of track data are included, the track chunk is separated into each track data. FIG. 3 shows an example in which the track chunk of the music file 4 is separated into m + 1 track data, that is, track 0 data 11-0, track 1 data 11-1, ..., Track m data 11-m. ing.

Hereinafter, with reference to the flowchart of FIG.
The flow of the track separation processing by the track separation unit 7 (including the file analysis unit 7a) will be described in more detail.

That is, when the track separation processing is started, first, the track separation unit 7 reads the information of the header chunk of the inputted music file 4 and the track recorded in a part of the header chunk. The number (m + 1 in FIG. 3) is substituted for the variable M (step S1). Then, the track separation unit 7 receives the input music file 4
Track chunks from the above track chunk
Data 11-0, track 1 data 11-1, ..., Track m data 11-m are sequentially read, and each is arranged in an array Trac.
The tracks are sequentially stored in k [0], Track [1], ..., Track [m] (steps S2 to S5), and the track separation unit 7 thus ends the track separation processing.

Next, the detailed configuration of the channel separation and channel number code generator 8 is as shown in FIG.
The channel separation / channel number code generation unit 8 has a track analysis unit 8a and a channel number code generation unit 8b.

The track analysis / separation unit 8a uses the track 0 data 11 previously divided by the track separation unit 7.
For each of the track data 0, track 1 data 11-1, ..., Track m data 11-m, it is analyzed whether or not track data related to a plurality of channels is included. , And separates each into track data consisting of a single channel. In the first embodiment, the track analysis / separation unit 8a converts the input track x data 11-x into track x data 12-0 only for channel 0 and track x data 12-only for channel 1.
, ..., Track x data 12-n of only channel n are separated. At this time, the channel number code generator 8b
The channel number code X13 of each data is generated. The channel number code X13 corresponds to information indicating attribution to the channel described in the claims.

Here, as described above, one track data can include a maximum of 16 channels in the SMF, but in the SMF format, an event belonging to a channel and an event not belonging to a channel (the whole is affected. Events etc.) can be described. In view of this, in the first embodiment, an event that does not belong to a channel is processed as channel 0. That is, an event that does not belong to a channel is included in the track x data of only channel 0. However, it is not limited to this.

In the first embodiment, if all the tracks in which no event exists are encoded, the number of tracks becomes huge. Therefore, the tracks in which no event exists are deleted and are not recorded in the primary code. There is.

Hereinafter, with reference to the flowchart of FIG.
The flow of processing relating to channel separation and channel number code generation by the channel separation and channel number code generation unit 8 (track analysis separation unit 8a, channel number code generation unit 8b) will be described in detail.

That is, when this processing is started, first, the channel separation / channel number code generation unit 8 substitutes 0 into the variable I relating to the track number (step S11),
This variable I is compared with the maximum track number M (step S12). In step S12, if the channel separation / channel number code generation unit 8 determines that the relationship of I <M is satisfied, the determination result in step S12 is Yes.
Proceeding to the s side, subsequently, 0 is assigned to the variable J relating to the channel number (step S13), and the variable J relating to this channel number and the maximum channel number MAX_CH are compared (step S1).
4).

Subsequently, in step S14, when the channel separation / channel number code generation section 8 determines that the relationship of J <MAX_CH is established, after setting Tick [J] to 0 (step S15), the channel is Variable J related to number
Is incremented by 1 (step S16), and the process returns to step S14. Here, Tick [J] means the elapsed time at the current coding position of the track data J being coded. Since each track has an independent time axis, such processing for the array Tick [J] is required. The channel separation / channel number code generation unit 8 executes the processes of steps S14 to S16 for all single channel track data separated from one track data.

For example, in the relationship with FIG. 5, since it is assumed that the track x data 11-x includes the event group belonging to channels 0 to n, the channel separation and channel number code generation unit. 8 is the above step S14
By repeating the processes from S16 to S16, Tick [0], Tic
Let k [1], ..., Tick [n] be 0.

After the above processing, if step S14 is advanced to No, the channel separation and channel number code generator 8
Substitutes 0 for the variable MasterTick and -1 for the variable Bch (step S17). This Bch is a variable indicating the channel number of the event immediately before (one before) the event being processed. Further, -1 assigned to Bch is a value indicating that the event being processed is at the head and that the immediately preceding event does not exist.

Then, it is determined whether or not there is an event in the track I data stored in Track [I] (step S18). The MasterTick means a variable relating to the elapsed time at the current purse position of the track data in which the event groups belonging to the plurality of separation source channels are stored.

Then, in step S18, the channel separation / channel number code generation unit 8 causes Track [I]
If it is determined that there is still an event in the track I data of, the Δtime and the event are read from the track I data, Δtime is stored in the variable Delta and the event is sequentially stored in the variable Event (step S20). Even
The channel number included in the event stored in t is stored in the variable Ch (step S21). Subsequently, the channel separation / channel number code generation unit 8 determines whether the Δtime stored in the variable Delta is 0, that is, whether the immediately preceding event is recorded at the same time (step S22). . And this step S2
In 2, when the channel separation and channel number code generation unit 8 determines that the Δtime is 0, the variable Bc
It is determined whether h is −1, that is, whether there is a previous event (step S23).

Then, in step S23, the channel separation and channel number code generator 8 sets the variable Bch to-.
1, that is, when it is determined that the immediately preceding event does not exist (the event being processed is at the head), step S24
Proceed to the subsequent processing. On the other hand, in step S23, when the channel separation / channel number code generation unit 8 determines that the variable Bch is not −1, that is, the event being processed is not the head, the variable is added to the end of the channel number code [I]. Encode the channel number assigned to Bch,
The process proceeds to step S24 and subsequent steps (step S27).

Subsequently, the channel separation / channel number code generation section 8 executes the three processes shown in the figure (step S
24), the value stored in the variable Cdelta and the event stored in the variable Event are added to the end of Track [I] [Ch] (step S25), and the channel number of the event immediately before the event being processed is added. After the channel number assigned to the variable Ch relating to the channel number currently being processed is assigned to the variable Bch relating to the sign (step S26), the process returns to step S18.

The channel separation / channel number code generation section 8 repeats the above processing as long as an event exists in the track I data of Track [I] (step S1).
8 to S26). When it is determined in step S18 that there is no other event in the track I data stored in Track [I], the variable I related to the track number is incremented by 1 (step S18).
19) and returns to step S12 to execute the same processing as described above for the next track data.

Then, the channel separation and channel number code generation unit 8 has the relationship of Track [m] in the relationship with FIG.
When the above-described processing is repeated up to the track m data 11-m stored in, the present processing ends. With such a series of processing, in relation to FIG. 5, the channel separation / channel number code generation unit 8 converts the input track x data 11-x into track x data 1 of channel 0 only.
2-0, track x data 12-1 of channel 1 only,
.. is separated into the track x data 12-n of only the channel n, and the channel number code X of each data is generated.

In the above process, the channel number of the immediately preceding event is encoded instead of the channel number of the event being processed so that the channel number code at the end of the event group at the same time is not encoded. Are doing

Here, with reference to FIG. 7, the processing relating to channel separation and channel number code generation by the channel separation and channel number code generation section 8 (track analysis separation section 8a, channel number code generation section 8b) described above is performed. Explain the effect.

Now, it is assumed that the track x data of a certain track x is constructed as shown in FIG. 7 (a). Since the track x data related to channels 4, 0, and 2 are mixed in the track x data illustrated in this example, the track x data for each channel is separated by the above-described processing. However, the state after separation is shown in FIG. That is, the upper part of FIG. 7B shows track x data consisting of a single channel 0, the middle part shows track x data consisting of a single channel 2, and the lower part shows track x data consisting of a single channel 4.

When decoding this to restore the original information, it is impossible to recognize the order in which the events A to C are arranged if only the information shown in FIG. 7B is used. I understand. For example, when the channels are arranged in the order of the channel numbers, the events B, C, and A are in the order, which is different from that before the track division.

In view of this point, in the first embodiment, the channel number code X is adopted so that the order of the events A, B and C can be recognized at the time of decoding. The channel number code X is coded in the manner as shown in FIG. 7C only when there are a plurality of events that occur at the same time. That is, in the order of the sequence of events, event A
A channel number for event B and a channel number for event B are encoded.

For example, in the example of FIG. 7A, since the Δtime of event B and event C is 0, it is known that these events A, B, and C have the same time, but three event A , B, and C, if the order of the two from the beginning is determined, it is understood that the order of the remaining one is unnecessary. Therefore, in the first embodiment, when there are n events at the same time, n-1 channel number codes are encoded.

That is, in the example of FIG. 7C, the channel number related to the event C is not encoded. More specifically,
When the primary code generation unit 2 determines that a plurality of events at the same time are recorded in the track data based on the information on the relative time between events (the first byte of Δtime), The event recorded at the end of the event group at the same time is excluded from the target holding the information indicating the attribution to the channel. The amount of data is also reduced by such a device.

It should be noted that if all the tracks in which no event exists are encoded, the number of tracks will be huge, so the first
In the embodiment of the present invention, the number of the tracks having no event is 1
It is deleted when the next code is generated and is not included in the primary code 5.

Next, the detailed construction of the code generator 9 other than the channel number is as shown in FIG. That is, the code generation unit 9 other than the channel number includes the event type analysis / identification unit 21, the Δ time code generation unit 22, and the Note
On pitch code generation unit 23, Note Off pitch code generation unit 2
4, velocity value code generation unit 25, differential bend value code generation unit 26, differential expression value code generation unit 27,
The event type identification code generation unit 28 and the other information code generation unit 29 are included.

Hereinafter, with reference to the flowchart of FIG.
Code generators 9 other than this channel number (each part 21 to 2
The flow of processing by (including 9) will be described in detail.

When this process is started, the code generator 9 other than the channel number first sets the variable I relating to the track number to 0.
Is set (step S31), and this variable I is compared with the maximum number of tracks M (m + 1 in the example of FIG. 3) (step S32).

Then, the code generator 9 other than the channel number
When it is determined that the relationship of I <M is established, 0 is substituted for the variable J related to the channel number (step S3
3) The variable J relating to the channel number is compared with a predetermined maximum number of channels MAX_CH (step S3).
4). In the first round of this loop, step S334 is performed first.
Since J = 0 in the above, the code generation unit 9 other than the channel number determines that the relationship of J <MAX_CH is satisfied, and the track 0 of only the channel 0 stored in Track [0] [0] is determined. The data is encoded (step S3)
6).

Then, the code generation unit 9 other than the channel number
After incrementing the variable J related to the channel number by 1 to J = 1, repeats the processes of steps S34 to S36 (step S37). Here, in the first embodiment, since the maximum number of channels MAX_CH is set to 16, the code generation unit 9 other than the channel number is
, Track [0] [0], Track [0] [1], ..., Track [0] [15] stored in Track [0] [1] ,. Only the track 0 data of 15 will be sequentially encoded.

Then, the code generator 9 other than the channel number
After the processing relating to the track 0 data, the variable I relating to the track number is incremented by 1 to set I = 1 (step S35), then Track [1] [0], Track [1]
[1], ..., Channel 0 stored in Track [1] [15]
Only track 1 data, channel 1 only track 1
Data, ..., Track 1 data of only channel 15 is to be sequentially encoded (step S32).
Through S37). Thus, the code generation unit 9 other than the channel number has Track [m] [0], Track [m] [1], ..., Track [m] [15].
When the track m data only for channel 0, the track m data only for channel 1, ..., The track m data only for channel 15 stored in the above are encoded by the same processing as described above (steps S32 to S3).
7) and this process ends.

Here, referring to the flowchart of FIG. 10, Track [I] executed in step S36 of FIG.
The encoding process of [J] will be described in more detail.

When the encoding process of Track [I] [J] is started, the code generator 9 for channels other than the channel number is in Track [I]
It is determined whether an event exists in the track I data of only the J channel stored in [J] (step S40). In step S40, if the code generation unit 9 other than the channel number determines that an event exists in the track I data of only the J channel, the Track
The Δtime and the event are read from the data of [I] [J], the Δtime is stored in the variable Delta, and the event is stored in the variable Event (step S41). Then, the code generation unit 9 other than the channel number determines the type of the event stored in this Event (step S42).

The discrimination of the event type discriminates the data string contained in the Event based on the SMF format (the SMF improved format in the second embodiment described later) specification. That is, in step S42, when the code generator 9 other than the channel number determines that the event stored in the variable Event is the event corresponding to type 1 in FIG. 79, the Δtime code
At the end of [I] [J], the Δtime stored in the variable delta of type 1 is encoded (step S43). On the other hand, in step S42, when the code generation unit 9 other than the channel number determines that the event stored in the variable Event is the event corresponding to type 2 in FIG. 79, the Δ time code. The delta time stored in the variable delta of type 2 is encoded at the end of [I] [J] (step S44).

On the other hand, in step S42, when the code generator 9 other than the channel number determines that the event stored in the variable Event is the event corresponding to type 3 in FIG. 79, the Δ time code. [I]
The delta time stored in the variable delta of type 3 is encoded at the end of [J] (step S45), and the code corresponding to the type of the variable Event is encoded at the end of the event type identification code [I] [J]. (Step S4)
6). In this way, a unique Δ that can identify the event type
By defining the time and performing encoding, step S42
When it is determined that the event is a type 1 or type 2 event, the event type identification code recorded in step S46 can be omitted, leading to a reduction in the amount of data.

Thus, although the details will be described later, the code generation unit 9 other than the channel number performs the coding for each event type (step S47), and the above processing (step S40).
Through S47) are repeated as long as there are events in Track [I] [J], and the above processing is performed for all events (step S40), the processing for encoding Track [I] [J] is completed. , And returns to step S36 in FIG.

Here, the primary code generation unit 2 calculates information on relative time between events, that is, the greatest common divisor of Δtime, and encodes the numerical value obtained by dividing the Δtime by the greatest common divisor. I am going to do it. That is, the primary code generation unit 2 encodes a value obtained by dividing the Δtime by the greatest common divisor when the Δtime is a variable length code. As a result, it is possible to reduce the capacity of the entire song.

Next, the encoding process for each event type executed in step S47 of FIG. 10 will be described in more detail with reference to the flowchart of FIG.

Now, when the coding for each event type is started, the code generator 9 for channels other than the channel number uses the variable Even
The type of the event stored in t is identified (step S50), and thereafter, the code generation unit 9 other than the channel number performs different encoding processing for each identified event (steps S51 to S58).

That is, the code generator 9 other than the channel number
If it is determined that the event type is Note On, the pitch value is encoded at the end of the Note On pitch code [I] [J] (step S51), and the velocity value code [I] [J] The velocity value is encoded at the end (step S52). on the other hand,
The code generation unit 9 other than the channel number
If it is determined to be Note Off, note off pitch code
The pitch value is encoded at the end of [I] [J] (step S5)
3). If the code generation unit 9 other than the channel number determines that the event type is pitch bend, it encodes the difference bend value from the immediately preceding bend value at the end of the difference bend value code [I] [J]. (Step S54).

The code generation unit 9 other than the channel number
If it is determined that the event type is expression, the code the difference expression value from the immediately preceding expression value at the end of the difference expression value code [I] [J] (step S55). When the code generation unit 9 other than the channel number determines that the event type is other, then it determines whether the event type identification code is 0x08 (see FIG. 21) or more. (Step S56).

Then, in step S56, the code generator 9 other than the channel number has the code 0x08 (see FIG. 21).
(Refer to) Other information code when it is judged as above
Information on the third and subsequent bytes of the event stored in the variable Event is encoded at the end of [I] [J] (step S5).
7). On the other hand, in step S56, the code generator 9 other than the channel number sets the code (hexadecimal number) to 0x08.
If it is determined that it is less than (see FIG. 21), the information after the second byte of the event stored in the variable Event is encoded at the end of the other information code [I] [J] (step S57). .

Thus, the code generation unit 9 other than the channel number ends the encoding process for each event type. still,
In the first embodiment, note off velocities are not generally used in the above-described encoding processing for each event type. Therefore, note off velocities are not encoded and are discarded, thereby reducing the amount of data. ing.

As described above, the code generation unit 9 (including the generation units 22 to 29) other than the channel number performs the Δ time code, Note On pitch code, N
The oteOff pitch code, velocity value code, differential bend value code, differential expression value code, event type identification code, and other information code are generated. Then, all these codes are input to the code arrangement unit 10 as shown in FIG. Then, the code arrangement unit 10 arranges each code type and generates the primary code 5.

Hereinafter, referring to FIGS. 13 to 23, the first
Code placement section 1 of performance information compression apparatus 1 according to the embodiment of
The flow of the code arrangement processing by 0 will be described in more detail.

Now, when this processing is started, the code arrangement unit 1
For 0, first, the code of the header chunk is arranged (step S60). After that, the arrangement of each code is executed.

First, the code arrangement unit 10 arranges the Δtime code (step S61).

That is, as shown in detail in the flowchart of FIG. 14, the code arrangement unit 10 first substitutes 0 into the variable I relating to the track number (step S70),
It is determined whether this I is smaller than the number M of tracks (step S71).

In the first round, I = 0 in step S70.
Is set, the code arrangement unit 10 determines that
From 71 proceed to the Yes side. Then, the code arrangement unit 10
Next, 0 is substituted for J related to the channel number (step S72).

Subsequently, the code arrangement unit 10 determines whether or not the variable J relating to this channel number is smaller than the maximum channel number MAX_CH (step S73).

Then, in one round, the above step S72
Since J = 0 is set, the code arrangement unit 10 proceeds from Step S73 to the Yes side. Then, the code arrangement unit 10 arranges the Δtime code [0] [0] (step S
74), the variable J related to the channel number is incremented by 1 to set J = 1 (step S75).

In the first embodiment, the maximum channel number MAX_CH is set to 16, so the code arrangement unit 10
Repeats the processing of steps S73 to S75,
Δ time code [0] [1], Δ time code [0] [2], ..., Δ time code [0] [15] are sequentially arranged. And Δ
After arranging the time codes [0] [15], the code arrangement unit 10
The process proceeds from step S73 to No, the variable I relating to the track number is incremented by 1 to set I = 1, then the processes of steps S71 to S75 are repeated again, and then Δ
Time code [1] [1], Δ time code [1] [2], ..., Δ time code [1] [15] are sequentially arranged. By the way, in this example, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG.

Therefore, the code arranging unit 10 repeats the processes of steps S71 to S75, and finally the Δ time code [m] [1] and Δ time code [m] [2] relating to the track m data. , .., .DELTA. Time codes [m] [15] are sequentially arranged, the delta time code arrangement processing is ended, and the process returns to step S61 in FIG.

Subsequently, the chord arranging section 10 lays out the Note On pitch code (step S62).

That is, as shown in detail in the flow chart of FIG. 15, the code arrangement unit 10 substitutes 0 for the variable I relating to the track number (step S80), and then this variable I is more than the number M of tracks. It is determined whether or not it is small (step S81).

In one round, I = 0 in step S80.
Is set, the code arrangement unit 10 determines that
From 81, go to the Yes side. Then, the code arrangement unit 10
Substitute 0 for the variable J related to the channel number (step S
82).

Subsequently, the code arrangement unit 10 determines whether or not the variable J relating to this channel number is smaller than the maximum channel number MAX_CH (step S83).

Then, in one round, the above step S82
Since J = 0 is set in step S11, the code arrangement unit 10 proceeds to Yes from step S83, and after arranging Note On pitch code [0] [0] (step S84), sets the channel number. The variable J is incremented by 1 to set J = 1 (step S85). In this first embodiment,
Since the maximum number of channels MAX_CH is set to 16, the code arranging unit 10 repeats the processes of steps S83 to S85 to repeat Note On interval code [0] [1], Note On interval code.
[0] [2], ..., Note On Pitch code [0] [15] are sequentially arranged.

Next, the code arranging section 10 arranges the Note On pitch codes [0] [15], and then No from step S83.
I will proceed to the side. Then, the code arrangement unit 10 increments I related to the track number by 1 to set I = 1, and then repeats the processing of steps S81 to S85, and then Note On pitch code [1] [1], Note On pitch code
[1] [2], ..., Note On pitch code [1] [15] are arranged in sequence.

By the way, in this example, the number of tracks M is set to m + 1 as shown in FIG. Therefore, the code arrangement unit 10 performs the steps S71 to S75.
Is repeated, and finally No related to track m data
te On pitch code [m] [1], Note On pitch code [m] [2], ..., No
te On When the pitch code [m] [15] is arranged sequentially, this Note On
The pitch code arrangement process is terminated, and the process returns to step S62 in FIG.

Subsequently, the code arranging section 10 arranges Note Off pitch codes (step S63).

That is, as shown in detail in the flowchart of FIG. 16, the code arrangement section 10 first substitutes 0 into the variable I relating to the track number (step S90),
It is determined whether this variable I is smaller than the number M of tracks (step S91).

First, in one round, since I = 0 is set in step S90, the code arranging unit 10 proceeds to Yes from step S91, and the code arranging unit 10 sets the variable J related to the channel number. 0 is substituted for (step S9
2).

Subsequently, the code arrangement unit 10 determines whether or not the variable J relating to this channel number is smaller than the maximum channel number MAX_CH (step S93).

Then, in one round, the above step S92
Since J = 0 is set in step S11, the code arrangement unit 10 proceeds from step S93 to the Yes side, and notes off the pitch code.
After arranging [0] and [0] (step S94), J related to the channel number is incremented by 1 to set J = 1 (step S95). Here, in the first embodiment, since 16 is set in advance in the maximum channel number MAX_CH,
After that, the code arrangement unit 10 repeats the processing of steps S93 to S95, and Note Off interval code [0] [1], Note Off interval code [0] [2], ..., Note Off interval code [0] [15]. Are sequentially arranged.

After arranging the Note Off pitch code [0] [15], the code arranging section 10 returns No from step S93.
After branching to the side and incrementing I for the track number by 1 to set I = 1, steps S91 to S95 are performed again.
Repeat the process of Note Off, pitch code [1] [1], Note Off
Pitch code [1] [2], ..., Note Off Pitch code [1] [15] are arranged sequentially. By the way, in this example, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG.

Therefore, the code arrangement unit 10 repeats the processing of steps S91 to S95, and finally the Note Off interval code [m] [1], Note Off interval code [m] [2] relating to the track m data. ], ..., When Note Off pitch codes [m] [15] are sequentially arranged, this Note Off pitch code arrangement processing is terminated, and the process returns to step S63 in FIG.

Subsequently, the code arrangement unit 10 arranges velocity value codes (step S64).

That is, as shown in detail in the flowchart of FIG. 17, the code arrangement unit 10 first substitutes 0 into the variable I relating to the track number (step S10).
0), it is determined whether or not this variable I is smaller than the number M of tracks (step S101).

In one round, I = I in step S100.
Since it is set to 0, the code arrangement unit 10 proceeds to Yes from step S101, and then substitutes 0 for J related to the channel number (step S102). Subsequently, the code arrangement unit 10 determines whether the variable J related to this channel number is smaller than the maximum channel number MAX_CH (step S103). Since J = 0 is set in step S102 in the first round, the code arrangement unit 10 proceeds to Yes from step S103, and the velocity value code
After arranging [0] [0] (step S104), J related to the channel number is incremented by 1 (step S10).
5). In the first embodiment, since MAX_CH related to the maximum number of channels is set to 16 in advance, the code arrangement unit 10 thereafter repeats the processes of steps S103 to S105 to repeat the velocity value code [0] [ 1], velocity value code [0] [2], ..., Velocity value code [0] [15] are sequentially arranged.

After arranging the velocity value codes [0] [15], the code arranging unit 10 executes the above step S10.
From No. 3, the No side is incremented, I for the track number is incremented by 1 to set I = 1, and then step S10 is performed again.
The processing from 1 to S105 is repeated, and subsequently, the velocity value code [1] [1], the velocity value code [1] [2], ..., The velocity value code [1] [15] are sequentially arranged. In this example, as shown in FIG.
It is assumed that +1 is set.

Therefore, the code arrangement unit 10 repeats the processing of steps S101 to S105, and finally the velocity value code [m] [1], the velocity value code [m] [2], ... , When the velocity value codes [m] [15] are sequentially arranged, this velocity value code arrangement processing is finished,
The process returns to step S64 in FIG.

Subsequently, the code arrangement unit 10 arranges the differential bend value code (step S65).

That is, as shown in detail in the flowchart of FIG. 18, the code arrangement unit 10 first substitutes 0 into the variable I relating to the track number (step S11).
0), it is determined whether or not this variable I is smaller than the number M of tracks (step S111).

In one round, I = I in step S110.
Since it is set to 0, the code arrangement unit 10 proceeds from step S111 to Yes. Then, the code arrangement unit 10
Substitutes 0 for the variable J related to the channel number (step S112).

Subsequently, the code arrangement unit 10 determines whether or not the variable J relating to this channel number is smaller than the maximum channel number MAX_CH (step S113).

In the first round, J = J in step S112.
Since it is set to 0, the code arrangement unit 10 proceeds from step S113 to Yes, arranges the differential bend value code [0] [0] (step S114), and then increments the variable J related to the channel number by one. Then, J = 1 is set (step S115). In the first embodiment, since MAX_CH related to the maximum number of channels is set to 16 in advance, the code arrangement unit 10 repeats the processing of steps S113 to S115, and the differential bend value code [0] [1]. , The differential bend value code [0] [2], ..., The differential bend value code [0] [15] are sequentially arranged.

After arranging the differential bend value codes [0] [15], the code arranging section 10 returns No to step S113.
To the side and increment the track number I by 1
= 1, the processes of steps S111 to S115 are repeated again, and then the differential bend value code [1] [1], the differential bend value code [1] [2], ..., The differential bend value code [1] [ 15] will be arranged sequentially. By the way, in this example, it is assumed that the number of tracks M is set to m + 1 as shown in FIG.

Therefore, the code arrangement unit 10 repeats the processing of steps S111 to S115, and finally the differential bend value code [m] [1] and the differential bend value code [m] [2] relating to the track m data. ,,, When the differential bend value codes [m] [15] are sequentially arranged, this differential bend value code arrangement processing is finished,
It returns to step S65 of FIG.

Next, the code arrangement unit 10 arranges the differential expression value code (step S66). That is, as shown in detail in the flowchart of FIG.
After substituting 0 for the variable I relating to the track number (step S120), the code arrangement unit 10 determines whether or not this variable I is smaller than the number M of tracks (step S12).
1).

In the first round, I = I in step S120.
Since it is set to 0, the code arrangement unit 10 proceeds to Yes from step S121, and then substitutes 0 into the variable J related to the channel number (step S122). Subsequently, the code arrangement unit 10 determines the variable J related to this channel number.
Is smaller than the maximum number of channels MAX_CH (step S123). In one round, the above step S1
Since J = 0 is set in step 22, the code arrangement unit 10
Proceeds from Step S123 to the Yes side.

Then, the code arrangement unit 10 arranges the differential expression value codes [0] [0] (step S12).
4), J related to the channel number is incremented by 1 to set J = 1 (step S125). In the first embodiment, since MAX_CH related to the maximum number of channels is set to 16 in advance, the code arrangement unit 10 repeats the processing of steps S123 to S125 thereafter, and the differential expression value code [0 ] [1], differential expression value code [0] [2], ..., differential expression value code
[0] [15] are sequentially arranged. Then, the code arrangement unit 10
After arranging the differential expression value code [0] [15], the process proceeds to No side from step S123, increments the variable I relating to the track number by 1 to set I = 1, and then performs the processing of steps S121 to S125 again. Then, the differential expression value code [1] [1], the differential expression value code [1] [2], ..., And the differential expression value code [1] [15] are sequentially arranged. By the way, in this example, as shown in FIG.
It is assumed that 1 is set.

Therefore, the code arranging section 10 repeats the processes of steps S121 to S125, and finally the differential expression value code [m] relating to the track m data.
When [1], the differential expression value code [m] [2], ..., The differential expression value code [m] [15] are sequentially arranged,
The differential expression value code arrangement processing is terminated, and
It returns to step S66 of 3.

Subsequently, the code arrangement unit 10 arranges the event type identification code (step S67). That is, FIG.
As shown in detail in the flowchart of 1., the code arrangement unit 10 first substitutes 0 into the variable I relating to the track number (step S130), and then determines whether the variable I is smaller than the number M of tracks. (Step S131).

In the first round, I =
Since the value is set to 0, the code arrangement unit 10 proceeds from step S131 to Yes and sets the variable J related to the channel number.
0 is substituted for (step S132). Subsequently, the code arrangement unit 10 determines whether or not the variable J related to this channel number is smaller than the maximum channel number MAX_CH (step S133). First, in one round, the above step S132
Since J = 0 is set in step S133, the code arrangement unit 10 proceeds to Yes from step S133, arranges the event type identification code [0] [0] (step S134), and then determines J related to the channel number. It is incremented by 1 to set J = 1 (step S135).

In the first embodiment, since MAX_CH related to the maximum number of channels is set to 16 in advance, the code arrangement unit 10 thereafter repeats the processing of steps S133 to S135, and thereafter, the event Type identification code
[0] [1], event type identification code [0] [2], ..., Event type identification code [0] [15] are sequentially arranged.

After arranging the event type identification codes [0] [15], the code arranging unit 10 proceeds to the No side from step S133, increments the variable I relating to the track number by 1 and sets I = 1. And then again in step S13
The processes from 1 to S135 are repeated, and subsequently the event type identification code [1] [1], the event type identification code [1] [2], ..., The event type identification code [1] [15] are sequentially arranged. By the way, in this example, as shown in FIG.
Is assumed to be set to m + 1.

Therefore, the code arrangement unit 10 repeats the processing of steps S131 to S135, and finally the event type identification code [m] [1] and the event type identification code [m] [2] relating to the track m data. ,…, Event type identification code [m]
When [15] is sequentially arranged, the event type identification code arrangement process is ended, and the process returns to step S67 in FIG.

The event type identification code is appropriately assigned according to the event type by the event type identification code encoding table as shown in FIG. In this example, 0x08 and above are assigned to events related to control changes.

Also, it is possible to dynamically allocate in accordance with the event type appearance frequency at the time of encoding. That is, in this case, the primary code generation unit 2 allocates different event type identification codes for each event included in the track data depending on the frequency of appearance, and collects the event type identification codes into independent areas. Deploy. It is needless to say that the compression efficiency by the Huffman coding at the time of the subsequent secondary coding can be improved even by the device at the time of the primary coding.

Subsequently, the code arrangement unit 10 arranges the channel number code (step S68).

That is, as shown in detail in the flowchart of FIG. 22, the code arrangement unit 10 first substitutes 0 into the variable I relating to the track number (step S14).
0), it is determined whether or not this variable I is smaller than the number M of tracks (step S141).

First, in one round, the above step S140.
Since I = 0 is set in step S1, the code arrangement unit 10 advances from step S141 to the Yes side, and the channel number code
After arranging [0] (step S142), the variable I relating to the track number is incremented by 1 to set I = 1 (step S143). Then, the code arrangement unit 10 repeats the processing of steps S141 to S143 again, and subsequently arranges the channel number code code [1]. by the way,
In this example, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG. Therefore, when the code arrangement unit 10 repeats the processing of steps S141 to S143 and finally arranges the channel number code code [m] relating to the track m data, the channel number code arrangement processing is ended, and the code number arrangement processing of FIG. Step S68
Return to.

Subsequently, the code arrangement unit 10 arranges other information codes (step S69).

That is, as shown in detail in the flowchart of FIG. 23, the code arrangement unit 10 first substitutes 0 into the variable I relating to the track number (step S15).
0), it is determined whether or not this variable I is smaller than the number M of tracks (step S151).

In the first round, I =
Since it is set to 0, the code arrangement unit 10 proceeds to Yes from step S151, and then substitutes 0 for J related to the channel number (step S152). Subsequently, the code arrangement unit 10 determines whether or not the variable J related to this channel number is smaller than the maximum channel number MAX_CH (step S153). And in one round, the above step S
Since J = 0 is set in 152, the code arrangement unit 10
Advances from step S153 to Yes, and after arranging the other information code [0] [0] (step S154), increments J related to the channel number by 1 to set J = 1 (step S155).

In the first embodiment, since MAX_CH related to the maximum number of channels is set to 16 in advance, the code arrangement unit 10 repeats the processing of steps S153 to S155 thereafter, and the other information code [0 ] [1], other information code [0] [2], ..., Other information code [0] [15] are sequentially arranged. Then, the code arrangement unit 10 causes the other information code [0] [1
5], the process proceeds to the No side from step S153, the variable I relating to the track number is incremented by 1 to set I = 1, and then steps S151 to S155 are performed again.
By repeating the above process, the other information code [1] [1], the other information code [1] [2], ..., And the other information code [1] [15] are sequentially arranged.

By the way, in this example, it is assumed that the number of tracks M is set to m + 1 as shown in FIG. Therefore, the code arrangement unit 10 repeats the processes of steps S151 to S155, and finally, the other information code [m] [1], the other information code [m] [2], ..., Other information related to the track m data. When the codes [m] [15] are sequentially arranged, the other information code arrangement processing ends. Thus
The code arrangement unit 10 ends all the processing related to code arrangement.

By the above processing, the codes are arranged in the basic arrangement order as shown in FIG. 24, and the primary code 5 is generated. That is, in more detail, in the example of FIG. 24, the basic arrangement order of the primary code 5 is Δ time code, Note On interval code, Note Off interval code, velocity value code, differential bend value code, differential expression value code. , Event type identification code, and other information code. Further, the internal arrangement of each code is, as shown in FIG. 25A, the order of codes relating to track 0 data (in order of channel number, the same hereinafter), track 1 data, ..., Track m data. . The internal arrangement order of channel number codes is as shown in FIG.

The primary code 5 itself is wasteful by the above-mentioned processing, and the music characteristic of the SMF is not impaired at all, resulting in a considerably compressed data amount.

The primary code 5 arranged in this way
Has a higher appearance frequency of the same data pattern than the SMF, the same data pattern is adjacent, and the probability that the same data pattern appears in the code of the same type is high. The efficiency of compression by the next code generation unit 3 is also increased.

The primary code 5 is input to the secondary code generation section 3, and the secondary code generation section 3 performs compression by the general-purpose compression method. In this example, the LZ coding method, the Huffman coding method, the run length coding method, and a method combining these (for example, LHA, ZIP, etc.) can be adopted as the general-purpose compression technology.

More specifically, assuming that the LZ encoding method is adopted, for example, first, the processing is started from the beginning of the primary code 5, and the data pattern of the processing target position and the previously processed predetermined range are set. Compare with the data pattern of. When the two match, the information such as the distance from the processing target position to the data pattern and the length of the matched data pattern is output as the secondary code 6. On the contrary,
If they do not match, the primary code 5 is output as it is as the secondary code 6.

Such a series of processing is repeated from the beginning to the end of the primary code 5.

In the first embodiment, as described above, since the primary code 5 in which each information is arranged in the region which is temporally independent for each information type and each channel is generated, the data pattern is repeated. This is also suitable for the LZ encoding method in which the repeated portion is compressed by utilizing this fact, and the compression rate can be lowered. Further, for example, with regard to the event type identification code, since it is possible to dynamically assign the code according to the frequency of occurrence of the event type, weighting is performed according to the frequency of occurrence of the code for each byte, and a small number of bits are assigned to a code having a weight. It is also suitable for the Huffman coding method for coding so that it can be recorded as a number, and the compression rate can be lowered. In addition to this, the velocity information is SMF.
Since the distribution of values is often biased under the format, recording the velocity information independently and collectively has a favorable effect on the efficiency of encoding by the Huffman encoding method.

In the first embodiment, it is assumed that the LZ method and the Huffman coding composite method (LHA) are adopted, but in that case, both effects described above are exhibited. Of course.

(Performance Information Decoding Device) Next, with reference to FIGS. 26 to 47, the performance information decoding device and the performance information decoding method according to the first embodiment for decoding the primary code or the secondary code. The performance information decoding program will be described in detail. The operation of the performance information decoding device corresponds to the performance information decoding method and the performance information decoding program according to the first embodiment.

First, FIG. 26 shows the configuration of a performance information decoding apparatus according to the first embodiment of the present invention, and will be described. As shown in FIG. 26, this performance information decoding device 50
Is composed of a secondary code decoding unit 51 and a primary code decoding unit 52. The primary code decoding means described in the claims corresponds to the primary code decoding section 52 and the like, and the secondary code decoding means described in the claims corresponds to the secondary code decoding section 51 and the like. However, it is not limited to this.

The secondary code decoding unit 51 converts the secondary code 6 compressed by the general-purpose compression method described above into the primary code 5
To decrypt. That is, in more detail, the primary code decoding unit 52 decodes the primary code 5 decoded by the secondary code decoding unit 51 into the music file 4 in the above-mentioned format. That is, when the secondary code 6 is decoded into the primary code 5 by the secondary code decoder 51, the primary code decoder 52
Thus, with respect to the primary code 5, each type of information arranged in an independent region is extracted for each type, and the extracted information is referenced for each channel while referring to the information indicating attribution to the channel. By being configured as a plurality of track data belonging to the above, and further configured as the track data of the file, the music file 4 is decoded.

Here, the detailed structure of the primary code decoding unit 51 is shown in FIG.

That is, the primary code decoding unit 51 has a channel code extraction unit 53, a track decoding unit 54, a track merge and channel number code decoding unit 55, and a track arrangement unit 56. Hereinafter, the configuration and operation of each of these units 53 to 56 will be described in detail.

First, the detailed configuration of the channel code extraction unit 53 is as shown in FIG. 28. The channel code extraction unit 53 extracts each code included in the primary code 5 for each channel. is there. That is, the code extraction unit 53 for each channel sequentially extracts the code group of track 0 and channel 0, the code group of track 0 and channel 1, ..., The code group of track m and channel n from the primary code 5 for each channel.
In addition, each code group includes Δ time code, Note On pitch code, N
Of course, the ote Off pitch code, velocity value code, differential bend value code, differential expression value code, event type identification code, and other information code are included.

Hereinafter, the flow of processing relating to the extraction of the code for each channel by the code for each channel extracting unit 53 will be described in detail with reference to the flowchart of FIG.

Now, when entering this processing, the channel-specific code extraction section 53 first extracts the code of the header chunk (step S160). After that, each code is extracted.

First, the channel code extraction unit 53 extracts the Δ time code (step S161). That is, FIG.
As shown in detail in the flowchart of 0, the channel-specific code extraction unit 53 first substitutes 0 into the variable I related to the track number (step S170), and then determines whether this variable I is smaller than the number M of tracks. Is determined (step S
171). In one round, I = 0 in step S170.
Is set, the channel-specific code extraction unit 53
The process proceeds from Step S171 to Yes, and then 0 is substituted into the variable J related to the channel number (Step S17
2).

Subsequently, the channel code extraction unit 53 determines whether or not the variable J relating to this channel number is smaller than the maximum channel number MAX_CH (step S173).

In one round, J = J in step S172.
Since it is set to 0, the channel-specific code extraction unit 53
Advances from Step S173 to the Yes side, extracts the Δ time code [0] [0] (Step S174), and then increments the variable J related to the channel number by 1 to set J = 1 (Step S175). In this first embodiment,
Since 16 is set in advance in MAX_CH related to the maximum number of channels, the channel-by-channel code extraction unit 53 repeats the processing of steps S173 to S175, and then the Δ time code [0] [1], Δ time code [ 0] [2], ..., Δ time code
[0] [15] are sequentially extracted.

Then, the channel code extraction unit 53
After the time code [0] [15] is extracted, the process proceeds to the No side from step S173, the variable I relating to the track number is incremented by 1, and then the above steps S171 to S175.
Repeat the process of Δ time code [1] [1], Δ time code
[1] [2], ..., ΔTime codes [1] [15] are sequentially extracted.

By the way, in the first embodiment, it is assumed that the number of tracks M is set to m + 1 as shown in FIG.

Therefore, the channel code extraction unit 53 repeats the processing of steps S171 to S175, and finally the Δ time codes [m] [1], Δ relating to the track m data.
When the time code [m] [2], ..., And the Δ time code [m] [15] are sequentially extracted, this Δ time code extraction processing is terminated, and
And returns to step S161.

Next, the channel-specific code extraction section 53
The On pitch code is extracted (step S162). That is,
As shown in detail in the flow chart of FIG. 31, the channel-specific code extraction unit 53 first substitutes 0 into the variable I relating to the track number (step S180), and then the variable I
Is smaller than the number M of tracks (step S181). First, in one round, the above step S18
Since 0 is set as I = 0, the channel code extraction unit 53 proceeds to Yes from step S181, and then substitutes 0 for J related to the channel number (step S1).
82).

Subsequently, the channel code extraction unit 53 determines whether or not the variable J relating to this channel number is smaller than the maximum channel number MAX_CH (step S183).

[0187] In the first round, J = J in step S182.
Since it is set to 0, the channel-specific code extraction unit 53
Proceeds from Step S183 to the Yes side, and after extracting Note On pitch code [0] [0] (Step S184), the variable J related to the channel number is incremented by 1 and J = 1.
(Step S185).

In the first embodiment, the maximum number of channels MAX_CH is set to 16 in advance, so that the channel-by-channel code extraction unit 53 will subsequently perform steps S183 to S18.
Repeat the process of 5 and repeat Note On pitch code [0] [1], Note On
Pitch code [0] [2], ..., Note On Pitch code [0] [15] are sequentially extracted. After extracting the Note On pitch code [0] [15], the channel code extraction unit 53 proceeds to No side from step S183, increments the variable I related to the track number by 1 and sets I = 1. After that, the processing of steps S181 to S185 is repeated again, and then the Note On pitch code
[1] [1], Note On pitch code [1] [2], ..., Note On pitch code
[1] [15] are sequentially extracted.

By the way, in this example, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG. Therefore, the channel-specific code extraction unit 53 repeats the processing of steps S181 to S185, and finally the Note On pitch code [m] [1] and Note On pitch code [m] [2] related to the track m data. , ..., Note On pitch codes [m] [15] are sequentially extracted, the Note On pitch code extraction processing is terminated, and the process returns to step S162 in FIG.

Next, the channel-specific code extraction section 53
The te Off pitch code is extracted (step S163). That is, as shown in detail in the flowchart of FIG.
The channel-specific code extraction unit 53 first substitutes 0 into the variable I relating to the track number (step S190), and then determines whether this variable I is smaller than the number M of tracks (step S191). On the first round, in step S190, I
= 0, the channel-specific code extraction unit 53
Proceeds from Step S191 to the Yes side, and then assigns 0 to the variable J related to the channel number (Step S19).
2).

Subsequently, the channel code extraction unit 53 determines whether or not the variable J relating to this channel number is smaller than the maximum channel number MAX_CH (step S193).

In the first round, J = J in step S192.
Since it is set to 0, the channel-specific code extraction unit 53
Proceeds from Step S193 to Yes, extracts Note Off interval code [0] [0] (Step S194), increments the variable J related to the channel number by 1 and J = 1.
(Step S195).

In this first embodiment, since MAX_CH related to the maximum number of channels is set to 16 in advance, the channel-specific code extraction unit 53 will thereafter perform the above step S1.
Repeat the processing from 93 to S195 to repeat Note Off pitch code
[0] [1], Note Off pitch code [0] [2], ..., Note Off pitch code [0] [15] are sequentially extracted.

After extracting the Note Off pitch code [0] [15], the channel-specific code extraction unit 53 proceeds to No side from step S193 and sets the variable I related to the track number.
Is incremented by 1 to set I = 1, and then the processes of steps S191 to S195 are repeated, and then Note
Off pitch code [1] [1], Note Off pitch code [1] [2], ..., No
te Off Pitch code [1] [15] is sequentially extracted.

By the way, in the first embodiment, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG. Therefore, the channel-specific code extraction unit 53 uses the steps S191 to S1.
The processing of 95 is repeated, and finally the Note Off pitch code [m] [1], Note Off pitch code [m] [2], which relates to the track m data,
…, Note Off interval codes [m] [15] are extracted sequentially, and this N
The ote Off pitch code extraction process is terminated, and the process returns to step S163 in FIG.

Then, the channel code extraction unit 53 extracts the velocity value code (step S164). That is, as shown in detail in the flowchart of FIG.
The channel-specific code extraction unit 53 first substitutes 0 into the variable I relating to the track number (step S200), and then determines whether this variable I is smaller than the number M of tracks (step S201). First, in one round, the above step S
Since I = 0 is set in 200, the channel-specific code extraction unit 53 proceeds to Yes from step S201, and then substitutes 0 into the variable J related to the channel number (step S202).

Subsequently, the channel code extraction unit 53 determines whether or not the variable J relating to this channel number is smaller than the maximum channel number MAX_CH (step S203).

In the first round, since J = 0 is set in step S202, the code extraction unit for each channel 5
3 proceeds to Yes from step S203, extracts velocity value code [0] [0] (step S204), and increments J associated with the channel number by 1 to set J = 1 (step S204). S205).

In the first embodiment, since MAX_CH related to the maximum number of channels is set to 16 in advance, the channel-by-channel code extraction unit 53 will thereafter perform the above step S2.
The processing of 03 to S205 is repeated, and the velocity value sign
[0] [1], velocity value code [0] [2], ..., Velocity value code [0] [15] are sequentially extracted.

After extracting the velocity value codes [0] [15], the channel-specific code extraction unit 53 executes step S2.
From 03, go to the No side and set the variable I related to the track number to 1
Is incremented by one to set I = 1, and then step S is performed again.
The processing from 201 to S205 is repeated, and subsequently, the velocity value code [1] [1], the velocity value code [1] [2], ..., The velocity value code [1] [15] are sequentially extracted.

By the way, in this example, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG. Therefore, the channel-specific code extraction unit 53 repeats the processing of steps S201 to S205, and finally the velocity value code [m] [1], the velocity value code [m] [2], ... , The velocity value code [m] [15] is sequentially extracted, the velocity value code extraction processing is ended, and the process returns to step S164 in FIG.

Subsequently, the differential bend value code is extracted (step S165).

That is, as shown in detail in the flowchart of FIG. 34, the channel code extraction unit 53 substitutes 0 into the variable I relating to the track number (step S21).
0), it is determined whether or not this variable I is smaller than the number M of tracks (step S211).

In the first round, I = I in step S210.
Since it is set to 0, the channel-specific code extraction unit 53
Proceeds from Step S211 to the Yes side, and then substitutes 0 for J related to the channel number (Step S21
2). Subsequently, the channel code extraction unit 53 determines whether or not the variable J related to this channel number is smaller than the maximum channel number MAX_CH (step S213). In the first round, since J = 0 is set in step S212, the channel-specific code extraction unit 53 proceeds from step S213 to the Yes side, and after extracting the differential bend value code [0] [0] (step S214), the variable J related to the channel number is set to 1
Is incremented by one (step S215).

In this first embodiment, since MAX_CH related to the maximum number of channels is set to 16 in advance, the channel-specific code extraction unit 53 will thereafter perform the above-mentioned step S2.
By repeating the processing from 13 to S215, the differential bend value code [0] [1], the differential bend value code [0] [2], ..., The differential bend value code [0] [15] are sequentially extracted.

After extracting the differential bend value codes [0] [15], the channel-specific code extraction unit 53 executes step S2.
From No. 13, proceed to the No side and set the variable I related to the track number to 1
Is incremented by one to set I = 1, and then step S is performed again.
The processing from 211 to S215 is repeated, and subsequently the differential bend value code [1] [1], the differential bend value code [1] [2], ..., The differential bend value code [1] [15] are sequentially extracted. Become.

By the way, in this example, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG. Therefore, the channel-specific code extraction unit 53 repeats the processes of steps S211 to S215, and finally the differential bend value code [m] [1] and the differential bend value code [m] [2] related to the track m data. ..., when the differential bend value codes [m] [15] are sequentially extracted, the differential bend value code extraction processing is ended, and the process returns to step S165 of FIG.

Subsequently, the channel code extraction unit 53 extracts the differential expression value code (step S1).
66). That is, as shown in detail in the flowchart of FIG. 35, the channel-specific code extraction unit 53 first substitutes 0 into the variable I relating to the track number (step S2).
20), it is determined whether or not this I is smaller than the number M of tracks (step S221). In the first round, since I = 0 is set in step S220, the channel-specific code extraction unit 53 advances to Yes from step S221, and then substitutes 0 into the variable J related to the channel number (step S222). .

Subsequently, the channel code extraction unit 53 determines whether or not the variable J relating to this channel number is smaller than the maximum channel number MAX_CH (step S223).

In the first round, J = J in step S222.
Since it is set to 0, the channel-specific code extraction unit 53
Proceeds from Step S223 to the Yes side, and after extracting the differential expression value code [0] [0] (Step S2
24), J related to the channel number is incremented by 1 to set J = 1 (step S225).

In this first embodiment, 16 is set in advance for MAX_CH related to the maximum number of channels, so that the channel-specific code extraction unit 53 hereafter performs the above step S2.
23 to S225 are repeated, and the differential expression value code [0] [1], the differential expression value code
[0] [2], ..., The differential expression value codes [0] [15] are sequentially extracted. After extracting the differential expression value code [0] [15], the channel-specific code extraction unit 53 proceeds to No side from step S123, increments the variable I related to the track number by one, and I = 1.
After that, the processing of steps S221 to S225 is repeated again, and then the differential expression value code [1]
[1], the differential expression value code [1] [2], ..., The differential expression value code [1] [15] are sequentially extracted. In this example, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG.

Therefore, the channel-specific code extraction unit 53 repeats the processing of steps S221 to S225, and finally the differential expression value code [m] [1] and the differential expression value code [m] [relating to the track m data. 2],
..., when the differential expression value codes [m] [15] are sequentially extracted, the differential expression value code extraction processing is terminated, and the process returns to step S166 in FIG.

Subsequently, the channel code extraction unit 53 extracts the event type identification code (step S167).
That is, as shown in detail in the flowchart of FIG. 36, the channel-specific code extraction unit 53 first substitutes 0 into the variable I relating to the track number (step S230), and then this variable I is smaller than the number M of tracks. It is determined whether or not (step S231). First, in one round, since I = 0 is set in step S230, the channel-specific code extraction unit 53 proceeds to Yes from step S231, and then substitutes 0 into the variable J relating to the channel number ( Step S232).

Subsequently, the channel-specific code extraction unit 53 determines that the variable J related to this channel number is the MA related to the maximum number of channels.
It is determined whether it is smaller than X_CH (step S23).
3).

In the first round, since J = 0 is set in step S232, the channel-specific code extraction unit 5
3 proceeds to Yes from step S233, and after extracting the event type identification code [0] [0] (step S22
4), J related to the channel number is incremented by 1 to set J = 1 (step S235).

In this first embodiment, since MAX_CH related to the maximum number of channels is preset to 16, the channel-specific code extraction unit 53 thereafter repeats the processing of steps S233 to S235 to identify the event type. Sign [0]
[1], event type identification code [0] [2], ..., Event type identification code [0] [15] are sequentially extracted.

After extracting the event type identification code [0] [15], the channel code extraction section 53 proceeds to No side from step S233 and sets the track number I
After incrementing by 1 to set I = 1, the processes of steps S231 to S235 are repeated again, and then the event type identification code [1] [1] and the event type identification code [1]
[2], ..., Event type identification codes [1] [15] are sequentially extracted.

By the way, in the first embodiment, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG. Therefore, the channel-specific code extraction unit 53 uses the steps S231 to S2.
The processing of step 35 is repeated, and finally the event type identification code [m] [1] and the event type identification code related to the track m data
[m] [2], ..., Event type identification codes [m] [15] are sequentially extracted, the event type identification code extraction processing is terminated, and
It returns to step S167 of 9.

Subsequently, the channel code extraction section 53 extracts the channel number code (step S168). That is, as shown in detail in the flowchart of FIG.
The channel-specific code extraction unit 53 first substitutes 0 into the variable I relating to the track number (step S240), and then determines whether or not this variable I is smaller than the number M of tracks (step S241). Then, in the first round, since I = 0 is set in step S240, the channel-specific code extraction unit 53 branches to Yes in step S241 and extracts the channel number code [0] (step S240). S
242), the variable I relating to the track number is incremented by 1 to set I = 1 (step S243).

Then, the channel code extraction unit 53 repeats the processing of steps S241 to S243 again, and subsequently extracts the channel number code code [1].

In this example, it is assumed that the number of tracks M is set to m + 1 as shown in FIG. Therefore, when the channel-specific code extraction unit 53 repeats the processing of steps S241 to S243 and finally extracts the channel number code [m] associated with the track m data, the channel-number code extraction processing ends, and Two
The process returns to step S168 of 9.

Subsequently, the channel-specific code extraction section 53 extracts other information codes (step S169). That is,
As shown in detail in the flowchart of FIG. 38, the channel-specific code extraction unit 53 first substitutes 0 into the variable I relating to the track number (step S250), and then determines whether this I is smaller than the number M of tracks. Is determined (step S
251). In one round, I = 0 in step S250.
Is set, the channel-specific code extraction unit 53
The process proceeds from Step S251 to Yes, and then 0 is substituted into the variable J related to the channel number (Step S25
2).

Then, the channel code extraction unit 53 determines whether or not the variable J related to this channel number is smaller than the maximum channel number MAX_CH (step S253).

Since J = 0 is set in step S252 in the first round, the code extraction unit for each channel 5
3 proceeds to the Yes side from step S253 and, after extracting the other information code [0] [0] (step S254), increments the variable J related to the channel number by 1 and J = 1.
(Step S255).

In the first embodiment, since MAX_CH related to the maximum number of channels is set to 16 in advance, the channel-by-channel code extraction unit 53 repeats the processing of steps S253 to S255, and then the other information code. [0] [1], other information code [0] [2], ..., Other information code [0] [15] are sequentially arranged.

Then, the code extraction unit 53 for each channel
After extracting the other information code [0] [15], the process proceeds to No side from step S253, and the variable I relating to the track number
Is incremented by 1, the processes of steps S251 to S255 are repeated again, and then the other information code
[1] [1], other information code [1] [2], ..., other information code
[1] and [15] will be sequentially extracted. By the way, in this example, as shown in FIG.
It is assumed that +1 is set.

Therefore, the channel-specific code extraction unit 53 repeats the processing of steps S251 to S255, and finally the other information code [m] [1] relating to the track m data,
When the other information code [m] [2], ..., And the other information code [m] [15] are sequentially extracted, the other information code extraction processing is ended, and thus, the code extraction for each channel is performed by the code extraction unit for each channel 53. The series of processes is ended. By the above processing, a code group consisting of a single channel (each code group includes Δ time code, Note On pitch code, Note Off pitch code, velocity value code, differential bend value code, differential expression value code, event type identification code , And other information codes) are extracted.

Next, the detailed structure of the track decoding unit 54 is as shown in FIG.

The track encoder 54 includes a Δ time code decoder 60, a Note On interval code decoder 61, a Note Off interval code decoder 62, a velocity value code decoder 63, a differential bend value code decoder 64, and a differential expression. Value code decoding unit 65, event type identification code decoding unit 66, other information code decoding unit 67, and event writing unit 68
have. Each of these units 60 to 67 decodes each code and causes the event writing unit 68 to write an event for each code, thus generating track data 69 consisting of a single channel.

The flow of processing relating to track decoding by the track decoding unit 54 will be described in detail below with reference to the flowchart in FIG.

When this process is started, the track decoding unit 5
4 first substitutes 0 into the variable I relating to the track number (step S260), and compares the variable I relating to this track number with the number M of tracks (step S261). Then, when the track decoding unit 54 determines that the relationship of I <M is established, 0 is substituted for the variable J related to the channel number (step S262), and the J related to the channel number is determined from the maximum channel number MAX_CH. Is also small (step S263).

Then, in the first round, the step S26 is performed first.
2, the track decoding unit 54 determines that the relationship of J <MAX_CH is satisfied, and Trac is Trac.
Decoding is performed on the track 0 data of only channel 0 stored in k [0] [0] (step S26).
5). Then, the track decoding unit 54 increments the variable J related to the channel number by 1 to set J = 1, and then
The processing of steps S263 to 266 is repeated (step S266).

In this first embodiment, MAX_CH = 16
Since this is predetermined, the track decoding unit 54
, Track [0] [0], Track [0] [1], ..., Track [0] [15] are stored in Track0 data only for channel 0, Track0 data only for channel1, ... , And decodes track 0 data of channel 15 only. Next, the track decoding unit 54 increments the track number I by 1, and then Track [1] [0], Track [1] [1], ..., Track [1] [1
5], track 1 data only for channel 0, track 1 data only for channel 1, ..., Track 1 data only for channel 15 are sequentially decoded (steps S261 to S266).

After that, the track decoding unit 54
[m] [0], Track [m] [1], ..., Track [m] [15] stored in Track [m] [15]
Only the track m data of only channel 15, ..., And the track m data of only channel 15 are decoded (step S26).
1 to S266), the present process ends.

Now, with reference to the flowchart of FIG. 41, the processing relating to the decoding of Track [I] [J] performed by the track decoding unit 54 in step S265 of FIG. That is, when this process is started, the track decoding unit 54 first determines whether or not a code exists in the Δtime code [I] [J] (step S270).

In step S270, when the track decoding unit 54 determines that a code exists in the Δtime code [I] [J], it reads the Δtime from the Δtime code [I] [J] and sets it as a variable. Store in Delta (step S27
1) After clearing the variable Event (step S27)
2) Decryption is performed for each event type, which will be described in detail later (step S273). Thus, the track decoding unit 54
Adds the Δtime related to the variable Delta and the event related to the variable Event in this order at the end of Track [I] [J], and returns to step S270 (step S274).

The track decoding unit 54 repeats the above-mentioned processing as long as there are codes in the delta time codes [I] [J], and when all the codes have been decoded, the process proceeds from Step S270 to the No side. The process ends.

Next, with reference to the flowchart in FIG. 42, the process relating to the decoding for each event type performed by the track decoding unit 54 in step S273 in FIG. 41 will be described in more detail. That is, when this process is started, the track decoding unit 54 determines the event type from the Δtime stored in the variable Delta (step S280).

As described above, in the first embodiment,
The type of event can be identified by the information of the first byte of Δtime. More specifically, in this example,
It is possible to identify three event types, namely, note-on (type 1), note-off (type 2), and other (type 3) events.

In step S280, when the track decoding unit 54 determines that the event is type 1, that is, Note On, the Note On status byte (J is assigned to the lower 4 bits in this example). Is generated and added to the end of the variable Event (step S281).
On Interval code [I] [J] read 1-byte code and variable
It is added to the end of Event (step S282), the 1-byte code is read from the velocity value code [I] [J], and the variable Ev
It is added to the end of ent (step S283).

On the other hand, in step S280, the track decoding unit 54 determines that the event is type 2, that is, Note O.
If it is determined to be ff, a Note Off status byte (J is assigned to the lower 4 bits in this example) is generated and added to the end of the variable Event (step S28).
4) Furthermore, a 1-byte code is read from Note On pitch code [I] [J] and added to the end of the variable Event (step S2).
85), the sign of the general Note Off velocity value is the variable Ev
It is added to the end of ent (step S286).

On the other hand, in step S280, when the track decoding unit 54 determines that the event is type 3, that is, other event, it sequentially reads one code from the event type identification codes [I] [J]. (Step S2
87) and subsequently, it is determined whether or not the event belongs to the channel (step S288).

When it is determined that the event belongs to the channel, the track decoding unit 54 determines the status byte (J in the lower 4 bits) according to the event.
Is substituted) and added to the end of the variable Event (step S289).

Subsequently, the track decoding section 54 determines whether or not the event type identification code is 0x04 (see FIG. 21) (step S290). In step S290, the track decoding unit 54 sets the event type identification code to 0x.
If it is determined that the value is 04, it can be seen from the table in FIG. 21 that the event is a pitch bend change, so a 2-byte code is read from the differential bend value code [I] [J],
The bend value added to the immediately preceding bend value is added to the end of the variable Event (step S291).

On the other hand, in step S290, if the track decoding unit 54 determines that the event type identification code is not 0x04 (see FIG. 21), then it determines whether it is 0x08 (see FIG. 21) or more. Yes (step S2
92).

Then, in this step S292,
When the track decoding unit 54 determines that the event type identification code is 0x08 or more, it recognizes from the table of FIG. 21 that the event is a control change event in which the control change number is omitted. A number is generated and added to the end of the variable Event (step S293).

Next, the track decoding section 54 determines whether or not the event type identification code is 0x08 (see FIG. 21) (step S294).

If it is determined in this step S294 that the event is 0x08, it can be seen from the table in FIG. 21 that the event is an expression, and therefore the 1-byte code from the differential expression value code [I] [J]. The expression value read out and added to the expression value immediately before is added to the end of the variable Event (step S295). On the other hand, if it is determined in step S292 that the event type identification code is not 0x08 or more, and if it is determined in step S294 that it is not 0x08,
The track decoding unit 54 sequentially reads the codes from the other information codes [I] [J] according to the SMF format as described at the beginning until the event is completed, and the variable Event
Is added to the end (step S297).

On the other hand, when the track decoding unit 54 determines in step S288 that the event does not belong to the channel, it generates a status byte corresponding to the event and adds it to the end of the event (step 296), the procedure moves to step S297. In this way, the track decoding unit 54 ends the series of processes related to the decoding for each event type.

Next, the detailed structure of the track merge and channel number code decoding unit 55 is as shown in FIG. As shown in the figure, the track merge section 55a
And a channel number code decoding unit 55b. Then, under such a configuration, the track merge and channel number code decoding unit 55 receives track x data only for channel 0, track x data only for channel 1, ..., Track x data only for channel n, and channel number code X. When input, the channel number code decoding unit 55b decodes the channel number code 13. Then, the track merge section 55a decodes the track x data 11-x while adding the channel number code. At this time, for events that do not belong to a channel, a predetermined channel (SMF is channel 0, and the improved format of SMF according to the second embodiment described later is used for channels 0 to 3 depending on the event class, as described in encoding. ) Is stored as an event belonging to (1), and is decoded in the same manner as an event belonging to a normal channel.

The flow of processing relating to track merging and decoding by the channel number code decoding unit 55 will be described below with reference to the flowchart in FIG.

When this process is started, the track merge and channel number code decoding unit 55 first substitutes 0 for I relating to the track number (step S300), and compares the variable I relating to this track number with the number M of tracks. Yes (step S301).

When the track merge and channel number code decoding unit 55 determines that the relationship of I <M is established, 0 is substituted into the variable MasterTick and the variable J related to the channel number (step S302). This variable MasterTick
Means the elapsed time (tick) at the current composite position of the track in which the event groups belonging to a plurality of channels after the recovery are stored in the track recovery at the time of the composite.

Next, the track merge and channel number code decoding unit 55 determines whether or not the variable J related to the channel number is smaller than the maximum channel number MAX_CH (step S303). In the first round, since J = 0 in step S302, the track merge / channel number code decoding unit 55 causes J <MAX_CH.
It is determined that the relationship of is satisfied, Tick [0] is set to 0, the variable J related to the channel number is further incremented, and J = 1.
After that (step S304), the above step S30
Return to 3. This Tick [x] means the elapsed time at the current purse position of the track where the event belonging to channel J is stored. Since each track has an independent time axis, such an arrangement is necessary.

In this example, since MAX_CH = 16 is predetermined, the track merge / channel number code decoding unit 55 executes the above processing for Tick [0], Tick [1], ..., Tick [15]. To do. After this process, the track merge and channel number code decoding unit 55 performs a track merge process which will be described in detail later (step S305) and increments the variable I by 1 to set I = 1 (step S30).
6) The above steps S301 to S306 are repeated, and in this way, track 1 data, track 2 data, ..., Track x, ..., Track m data are decoded.

Here, referring to the flow chart of FIG. 45, in step S305 of FIG. 44, the track merge and channel number code decoding unit 55 (track merge unit 55
a, including the channel number code decoding unit 55b), the process relating to track merging will be described in more detail. When this process is started, the track merge and channel number code decoding unit 55 causes each variable NextTick, Ex used in this process.
istSameTimeEvent, after initializing the variable J (step S
310), this variable J is compared with the maximum channel number MAX_CH (step S311). In the first round, step S31
Since the variable J is set to 0 in 0, the track merge and channel number code decoding unit 55 proceeds to Yes from step S311 and then Track [I].
It is determined whether the reading of [0] is completed (step S312).

Here, since the reading of Track [I] [0] is not completed, the track merging / channel number code decoding unit 55 proceeds to No side from step S312 and does not advance the reading position, and thus Track [I] ] [0] Track data Δ
The time is read and stored in the variable Delta (step S
313). Subsequently, the track merge and channel number code decoding unit 55 determines whether or not NextTick = Tick [J] + Delta (step S314), and NextTick = Tick [J].
If + Delta, ExistSameTimeEvent = 1 is set (step S315). If NextTick = Tick [J] + Delta is not satisfied, it is determined whether or not NextTick> Tick [J] + Delta (step S316).

When this relationship is established, the track merge and channel number code decoding section 55 determines that NextTick = Ti
After setting ck [J] + Delta, ExistSameTimeEvent = 0, and Ch = J (step S317), the variable J is incremented by 1 to J = 1 (step S318). And
The track merging / channel number code decoding unit 55 performs the process of steps S311 to S318 by Track [I].
[0], Track [I] [1], ..., Track [I] [15] are repeated.

When the above step S311 proceeds to the No side,
The track merge and channel number code decoding unit 55 uses the Ne
It is determined whether xtTick = 0xffffffff (step S
319), if it is determined that the relationship of NextTick = 0xffffffff is not established, it is determined whether ExistSameTimeEvent = 1 (step S320).

Then, in step S320,
The track merge and channel number code decoding unit 55
When it is determined that istSameTimeEvent = 1, one code is read from the channel number code [I] and is substituted for Ch (step S321). Further, from Track [I] [Ch] to Delta to Δtime and Event to Event. Is read (step S3
22). On the other hand, in step S320, the track merge and channel number code decoding unit 55 causes the ExistSameTim
If it is determined that eEvent = 0, Track [I] [Ch] to De
Read the Δtime in lta and the event in Event (step S323), and move from Track [I] [Ch] to Mdel without advancing the read position.
The next Δtime is read into ta (step S324), Md
It is determined whether or not elta = 0 (step S32).
5).

If Mdelta = 0, the track merge and channel number code decoding unit 55
The channel number coding [I] or al 1 codes for reading empty (step S326). Next, the track merge / channel number code decoding unit 55 updates and encodes the tick relating to the track data of the channel from which the event is read Δ
The time is calculated and assigned to Mdelta, and MasterTick is updated (step S327). Then, the Δtime assigned to Mdelta and the event assigned to Event are added to the end of Track [I] (step S328), and the above step S is performed.
Returning to 310, the above processing is repeated. Then, in step S319, the track merge and channel number code decoding unit 55 has NextTick = 0xffffff.
When it is determined that the relationship of ff is established, this process ends. In this way, track 0 data, track 1 data, ..., Track m data for each track is generated.

These are input to the track arrangement unit 56, arranged in tracks by the track arrangement unit 56, and thus decoded into the music file 4.

The processing relating to the track arrangement by the track arrangement unit 56 will be described in detail below with reference to the flowchart in FIG. That is, when this process is started, the track arrangement unit 56 first arranges a header chunk (step S350), then initializes I relating to the track number to 0 (step S351), and sets a variable I relating to this track number. And the number M of tracks are compared (step S35).
2). Then, when the relationship of I <M is established, the track arrangement unit 56 arranges the track 0 data stored in Track [0] (step S353) and increments I related to the track number to I. After setting = 1, the processes of steps S352 to 354 are repeated.

In the first embodiment, since M = m + 1 in the relationship with FIG. 46, the track arrangement section 56.
... arranges track 0 data, track 1 data, ..., Track m data stored in Track [0], Track [1], ..., Track [m] (step S35).
2 to S354), and this processing ends.

By the series of processing described above, 2
The primary code 5 is decoded by the secondary code 6 to the secondary code decoding unit 51, and the music file 4 is decoded by the primary code 5 to the primary code decoding unit 52. The musical composition file 4 thus decoded completely matches the original performance information without changing the order of the performance information.

As described above, the performance information compression apparatus, the performance information compression method, the performance information compression program, the performance information decoding apparatus, the performance information decoding method, and the performance information decoding program according to the first embodiment of the present invention are The following effects are exhibited.

That is, in the performance information compression device and the like, the music file 4 (SMF) is processed by the processing by the primary code generator 2.
1 is set as the compressed data amount without waste of data and without degrading the performance information of the music file 4 (SMF).
It realizes to generate the next code 5. Furthermore, this 1
The next code 5 has a higher occurrence frequency of the same data pattern than the SMF, the same data pattern is recorded at a closer distance, and the same code bytes are grouped at a closer position, so that the secondary code is generated. The compression efficiency of the general-purpose compression method in the code generation unit 3 is improved.

In the performance information decoding device and the like, this effect is similarly exhibited in the primary code decoding unit 52 and the secondary code decoding unit 51. Moreover, by preserving the information relating to the order of each performance information while temporally making each performance information independent at the time of the primary encoding, the order of each performance information is decoded when the encoded performance information is decoded. Does not change. That is, the decrypted performance information completely matches the original performance information.

(Second Embodiment) First, a performance information compression apparatus, a performance information decoding apparatus, a performance information compression method, a performance information decoding method, a performance information compression program, and a performance information decoding apparatus according to the second embodiment. The format of input data (music file) that is assumed in the program will be described in detail. In the second embodiment, an improved SMF format is assumed as the music file format. This improved format is
It is roughly classified into a header chunk and a track chunk. Each header chunk portion contains minimum information directly related to the entire music file, for example, information such as the number of track data. And for each track chunk,
Event data groups (track data) relating to a maximum of 4 channels can be included.

Further, each track data is composed of two elements such as Δtime and event. The Δtime includes at least information such as a relative time between adjacent events. In the SMF described above, the event includes the note-on or note-off status, pitch (note number), and performance information of the sound intensity (velocity), but note-off is omitted in this improved format. However, it differs from SMF in that the note message corresponding to note-on also includes information on "sound length".

Also in the second embodiment, the type of event can be identified by the information of the first byte of Δtime. Also, in this format, the status may include event class information.

Due to the difference in the format as described above, the music file, the primary code, and the secondary code are different from those according to the first embodiment. The same reference numerals as those used in the embodiment are used for description.

(Performance Information Compressing Device, etc.) With reference to FIGS. 48 to 61, a performance information compressing device according to the second embodiment for generating a primary code and further a secondary code from a music file. The configuration and operation of the performance information compression method and the performance information compression program will be described in detail.

The operation of this performance information compression device is as follows.
It corresponds to a performance information compression method and a performance information compression program according to the second embodiment. The performance information compression program and the performance information decoding program can be executed by a computer.

Further, here, in order to avoid duplicated description, the same configuration as that of the apparatus and the like according to the first embodiment will be described with reference to the same drawings.

The configuration of the performance information compression apparatus 70 according to the second embodiment is substantially the same as the configuration shown in FIG. 1 described above, and is composed of the primary code generation section 2 and the secondary code generation section 3. Become.

That is, the primary code generator 2 rearranges the performance information contained in the music file 4 in the above-mentioned format in order to improve the compression efficiency in the secondary code generator 3, and
The next code 5 is generated. The secondary code generation unit 3 compresses the primary code 5 by a general-purpose compression method to generate a secondary code 6.

Also in the second embodiment, as the general-purpose compression method adopted in the secondary code generation section 2, for example, LZ
Encoding method, Huffman encoding method, run-length encoding method,
Furthermore, a method that combines these (for example, LHA or ZIP
Etc.) is adopted, but the invention is not limited to these.

The detailed configuration of the primary code generation unit 2 is substantially the same as the configuration shown in FIG. 2 described above, and the track separation unit 7, the channel separation and channel number code generation unit 8, and the code generation other than the channel number. It has a unit 100 (described later in FIG. 48) and a code arrangement unit 10.

The configuration and operation of each of these parts will be described in detail below.

First, the detailed structure of the track separating section 7 is as follows.
The configuration is similar to that shown in FIG. 3 and has a file analysis unit 7a. The file analysis unit 7a determines whether or not the track chunk of the music file 4 includes a plurality of track data based on the information on the number of track data included in the header chunk of the music file 4. Then, when it is determined that a plurality of track data are included, the track separating unit 7 separates the respective track data.

In this example, the track chunks are track 0 data 11-0, track 1 data 11-1,
..., track m data 11-m are separated. Since the track separation processing by the track separation unit 7 is substantially the same as that shown in the flow chart of FIG. 4, duplicate description will be omitted here.

As described above, in the second embodiment, since the input data (music file) of the improved format different from that of the first embodiment is assumed, the separated track data is obtained. , There are some differences from the above-described first embodiment, but here, in order to simplify the description, the track chunks are track 0 data 11-0, track 1 data 11-1 ,. The description will proceed with the same reference numerals as the track m data 11-m.

Next, the detailed configuration of the channel separation / channel number code generation section 8 is substantially the same as the configuration shown in FIG. 5, and has a track analysis section 8a and a channel number code generation section 8b. ing. In the case where each of the track 0 data 11-0, the track 1 data 11-1, ..., The track m data 11-m divided by the track separation unit 7 as described above includes track data related to a plurality of channels. The channel analysis / separation unit 8a of the channel separation / channel number code generation unit 8 separates each into track data consisting of a single channel.

More specifically, in the performance information compression apparatus and the like according to the second embodiment, the track analysis / separation unit 8a
Divides track x data 11-x into track x data 12-0 only for channel 0, track x data 12-1 only for channel 1, ..., Track x data 12-n only for channel n. At this time, the channel number code generator 8b generates these channel number codes X13.

As described above, the improved format can include a maximum of 4 channels in one track data. However, in the improved format, events belonging to the channel and events not belonging to the channel (impacted to the whole are affected). It is possible to describe the event that covers). In view of this, in the second embodiment, an event that does not belong to a channel is treated as belonging to each channel according to the event class. In this example, the class of event defines the channel to store as follows:

That is, in the case of the normal class, the stored channel is channel 0, in the case of class A the stored channel is channel 1, in the case of class B the stored channel is channel 2, and in the case of class C. The channel to be stored will be defined as channel 2. Then, it is included in the track x data of only each channel based on the above event class information. However, it is not limited to this. Also in the second embodiment, as in the first embodiment, if all the tracks in which no event exists are encoded, the number of tracks becomes huge, so the tracks in which no event exists are deleted. It is not recorded in the next code.

As described above, in the second embodiment, since the input data (music file) of the improved format different from that of the first embodiment is assumed, the separated track data is obtained. Is slightly different from that of the first embodiment, but here, in order to simplify the explanation, the track x data 12-x, the track 0 channel only x data 12-0, and the channel 1 only track are shown. .., track n data only for channel n, x data 12-n, track 1 data 11-1, ..., Track m data 11-m.

The flow of processing relating to channel separation and channel number code generation by the track analysis / separation unit 8a and channel number code generation unit 8b is shown in the flow chart of FIG. 6 except that MAX_CH = 4 is set. Since it is substantially the same as the one described above, the duplicate description is omitted here. Further, the effects of the processing relating to the channel separation and the channel number code generation by the track analysis / separation unit 8a and the channel number code generation unit 8b are the same as the contents described above with reference to FIG.

Next, the code generator 10 other than the channel number
The detailed configuration of 0 is as shown in FIG. That is, the code generation unit 100 other than the channel number includes the event type analysis / identification unit 101 and the Δ time code generation unit 10.
2, pitch code generation unit 103, gate time code generation unit 1
04, velocity value code generation unit 105, difference bend value code generation unit 106, difference volume value code generation unit 107,
The event type identification code generation unit 108 and the other information code generation unit 109 are included.

The code generation unit 9 other than this channel number
Since the flow of the processing by is almost the same as the content described above with reference to the flowchart of FIG. 9 except that MAX_CH = 4, duplicate description will be omitted. Further, the processing of encoding Track [I] [J] executed in step S36 of FIG. 9 is substantially the same as the content described above with reference to the flowchart of FIG. Omit it.

However, the encoding process for each event type executed by the code generation unit 9 other than the channel number in step S47 of FIG. 10 is different from that of the first embodiment, and therefore, hereinafter, FIG. The detailed description will be made with reference to the flowchart of FIG. 1 while being aware of the difference from the first embodiment described above.

When this process is started, the code generator 9 other than the channel number first determines the type of event stored in the variable Event (step S360).

If the code generator 9 other than the channel number determines that the event type is a note message, it encodes the lower 6 bits of velocity information at the end of the velocity value code [I] [J]. (Step S361), the gate time is encoded at the end of the gate time code [I] [J] (step S362), and the pitch value (key shift to upper 2 bits, lower 6 bits) at the end of the pitch code [I] [J] Key number) is encoded (step S363), and other information code
At the end of [I] [J], the information from the 4th byte onward of the event stored in the variable Event is encoded (step S37).
0), the process ends.

Then, the code generation unit 9 other than the channel number
If it is determined that the event type is the gate time extension, the above steps S362 to S362 are performed.
The same processing as 370 is performed, and this processing ends. Further, when the event other than the channel number is determined to be the channel pitch bend, the code generation unit 9 encodes the differential pitch bend from the immediately preceding pitch bend at the end of the differential bend value code [I] [J] (step (S364), this process ends. When the event other than the channel number is determined to be the channel volume, the code generation unit 9 encodes the differential volume from the immediately preceding volume at the end of the differential volume value code [I] [J] (step S364), this processing is to be ended.

On the other hand, the code generator 9 other than the channel number
When it is determined that the event is of another type (step S360), first, the event type identification code is 0x.
It is determined whether it is 08 (see FIG. 58) or more (step S366).

In step S366, the code generator 9 other than the channel number sets the event type identification code to 0x.
If it is determined that the value is 08 (see FIG. 58) or more, the third byte information of the event stored in the variable Event is added to the end of the other information code [I] [J] (at this time, the channel index is Encoded (cleared to 0) (step S
371), and this processing ends. On the other hand, the above step S
If it is determined in 366 that the event type identification code is less than 0x08, the event type identification code is further 0x02.
(See FIG. 58) It is determined whether or not the following (step S367).

In step S367, the code generator 9 other than the channel number sets the event type identification code to 0x.
If it is determined that the value is 02 (see FIG. 58) or less, the second byte information of the event stored in the variable Event is encoded at the end of the other information code [I] [J], and the other information code [ Information on the third byte of the event stored in the variable Event is encoded at the end of I] [J] (step S36).
9) Furthermore, the information from the 4th byte onward of the event is encoded at the end of the other information code [I] [J] (step S36).
9), this processing is to be ended.

On the other hand, in step S367, if the code generator 9 other than the channel number determines that the event type identification code is not 0x02 (see FIG. 58) or less, the other information code [I] Variable Event at the end of [J]
The information of the third byte of the event stored in is encoded (step S369), and the information of the fourth byte and later of the event stored in the variable Event is encoded at the end of the other information code [I] [J]. Then (step S369), this process ends.

Each of the generating units 102 to 1 as described above
By a series of processing by 09, the Δ time code, the pitch code, the gate time code, the velocity value code, the difference bend value code, the difference volume value code, the event type identification code, and the other information code are generated. And
All of these codes are input to the code arranging unit 10 that is substantially the same as that shown in FIG. The code arrangement unit 10 arranges each code type and generates a primary code 5. The code arrangement unit 10 also has a slightly different configuration from the first embodiment described above, but here, in order to simplify the description,
The same reference numerals will be used to explain in detail focusing on the difference in processing.

The flow of code arrangement processing by the code arrangement unit 10 of the performance information compression apparatus 70 according to the embodiment will be described in more detail below with reference to FIGS. 50 to 60.

Now, when entering this processing, the code arrangement unit 10
First arranges the code of the header chunk (step S380). Hereinafter, each code is arranged as described later.

First, the code arrangement unit 10 arranges the Δtime code (step S380).

That is, as shown in detail in the flowchart of FIG. 51, the code arrangement unit 10 first substitutes 0 into the variable I relating to the track number (step S39).
0), it is determined whether or not this I is smaller than the number M of tracks (step S391).

In one round, I = I in step S390.
Since it is set to 0, the code arrangement unit 10 proceeds to Yes from step S391, and then substitutes 0 for J related to the channel number (step S392). Subsequently, the code arrangement unit 10 determines whether the variable J related to this channel number is smaller than the maximum channel number MAX_CH (step S393). And in one round, the above step S
Since J = 0 is set in 392, the code arrangement unit 10
Proceeds from Step S393 to the Yes side, and Δ time code
After arranging [0] [0] (step S394), the variable J related to the channel number is incremented by 1 to set J = 1 (step S395).

In this second embodiment, since MAX_CH relating to the maximum number of channels is set to 4 in advance, the code arrangement unit 10 repeats the processing of steps S393 to S395, and the Δ time code [0 ] [1], Δ time code [0]
[2], Δ time code [0] [3], and Δ time code [0] [4] are sequentially arranged. The code arrangement unit 10 uses the Δ time code [0] [4].
After arranging, the process proceeds to the No side from step S393, increments the variable I related to the track number by 1 to set I = 1, and then again performs steps S391 to S395.
Then, the Δ time code [1] [1], Δ time code [1] [2], Δ time code [1] [3], Δ time code [1]
Place [4] sequentially.

Also in this example, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG. Therefore, the code arrangement unit 10 causes the above-mentioned step S3.
The processing from 91 to S395 is repeated, and finally the Δ time code [m] [1] and Δ time code [m] relating to the track m data
When [2], Δ time code [m] [3], and Δ time code [m] [4] are sequentially arranged, this delta time code arrangement processing is finished,
It returns to step S381 of FIG.

Subsequently, the code arrangement unit 10 arranges the pitch code (step S382).

That is, as shown in detail in the flowchart of FIG. 52, the code arrangement unit 10 first sets 0 to the variable I relating to the track number (step S400), and then this variable I is smaller than the number M of tracks. It is determined whether or not (step S401).

In the first round, I =
Since it is set to 0, the code arrangement unit 10 proceeds to Yes from step S401, and substitutes 0 into the variable J related to the channel number (step S402). Subsequently, the code arrangement unit 10 determines whether or not the variable J related to this channel number is smaller than the maximum channel number MAX_CH (step S403). Then, in one round, the above step S4
Since J = 0 is set in 02, the code arrangement unit 10
Goes to the Yes side from step S403, and the pitch code
[0] After arranging [0] (step S404), the variable J related to the channel number is incremented by 1 (step S405).

In this second embodiment, since MAX_CH related to the maximum number of channels is preset to 4, the code arrangement unit 10 thereafter performs the above steps S403 to S405.
Repeat the process of pitch code [0] [1], pitch code [0] [2],
The pitch code [0] [3] and the pitch code [0] [4] are sequentially arranged. In this way, the code arrangement unit 10 proceeds to No side from step S403 after the pitch code [0] [4] is arranged, increments I related to the track number by 1 and sets I = 1,
The processing of steps S401 to S405 is repeated again,
Next, pitch code [1] [1], pitch code [1] [2], pitch code [1]
[3] and pitch codes [1] and [4] will be sequentially arranged. In this example as well, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG.

Therefore, the code arrangement unit 10 repeats the processing of steps S401 to S405, and finally the pitch code [m] [1], the pitch code [m] [2], and the pitch code [m] [2] relating to the track m data.
When the pitch code [m] [3] and the Note On pitch code [m] [4] are sequentially arranged, this pitch code arrangement processing is ended, and the process returns to step S382 of FIG.

Subsequently, the code arrangement unit 10 arranges the gate time code (step S383).

That is, as shown in detail in the flowchart of FIG. 53, the code arrangement unit 10 first substitutes 0 into the variable I relating to the track number (step S410), and then this variable I is smaller than the number M of tracks. It is determined whether or not (step S411).

In one round, I = I in step S410.
Since it is set to 0, the code arrangement unit 10 proceeds to Yes from step S411, and sets 0 to the variable J related to the channel number (step S412). Subsequently, the code arrangement unit 10 determines whether the variable J related to this channel number is smaller than MAX_CH related to the maximum number of channels (step S413). Since J = 0 is set in step S412 in the first round, the code arrangement unit 10 advances to Yes from step S413,
After arranging the gate time codes [0] [0] (step S41
4), J related to the channel number is incremented by 1 (step S415).

In this second embodiment, since MAX_CH relating to the maximum number of channels is preset to 4, the code arrangement unit 10 thereafter repeats the processing of steps S413 to S415, and the gate time code [0 ] [1], gate time code [0] [2], gate time code [0] [3], and gate time code [0] [4] are sequentially arranged.

After the gate time codes [0] [4] are arranged, the code arrangement unit 10 proceeds to step S413 to No side and increments I relating to the track number by 1 to set I = 1. After that, the processes of steps S411 to S415 are repeated again, and subsequently, the gate time code [1] [1], the gate time code [1] [2], the gate time code [1] [3], and the gate time code [1]. Place [4] sequentially. In this example as well, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG. Therefore, the code arrangement unit 1
0 repeats the processing of steps S411 to S415, and finally the gate time code [m] [1], the gate time code [m] [2], and the gate time code related to the track m data.
When [m] [3] and the gate time code [m] [4] are sequentially arranged, this gate time code arrangement processing is ended, and the process returns to step S383 in FIG.

Subsequently, the code arrangement unit 10 arranges velocity value codes (step S384).

That is, as shown in detail in the flowchart of FIG. 54, the code arrangement unit 10 first substitutes 0 into the variable I relating to the track number (step S420), and then this variable I is smaller than the number M of tracks. It is determined whether or not (step S421).

In one round, I = I in step S420.
Since it is set to 0, the code arrangement unit 10 proceeds to Yes from step S421, and then substitutes 0 for J related to the channel number (step S422). Subsequent to this, the code arranging unit 10 sets the variable J relating to this channel number.
Is smaller than MAX_CH related to the maximum number of channels (step S423). Since J = 0 is set in step S422 in one round, the code arrangement unit 10 proceeds to Yes from step S423 and arranges velocity value codes [0] [0] (step S42).
4), J related to the channel number is incremented by 1 (step S425).

Also in this second embodiment, since MAX_CH related to the maximum number of channels is set to 4 in advance, the code arrangement unit 10 thereafter performs the above steps S423 to S4.
The processing of 25 is repeated to sequentially arrange the velocity value code [0] [1], the velocity value code [0] [2], the velocity value code [0] [3], and the velocity value code [0] [4].

After arranging the velocity value code [0] [4], the code arranging unit 10 returns No from step S423.
To the side, I is incremented by 1 for the track number to set I = 1, and then steps S421 to S42 are performed again.
Repeat the process of 5 and then the velocity value sign
[1] [1], velocity value sign [1] [2], velocity value sign
[1] [3] and velocity value codes [1] [4] are sequentially arranged.

Also in this example, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG. Therefore, the code arrangement unit 10 performs the above step S4.
The processing from 21 to S425 is repeated, and finally the velocity value code [m] [1], the velocity value code [m] [2], the velocity value code [m] [3], the velocity value code [ When m] [4] are sequentially arranged, this velocity value code arrangement processing ends, and the process returns to step S384 in FIG.

Subsequently, the code arrangement unit 10 arranges the differential bend value code (step S385).

That is, as shown in detail in the flow chart of FIG. 55, the code arrangement unit 10 first substitutes 0 into the variable I relating to the track number (step S430), and then this variable I is smaller than the number M of tracks. It is determined whether or not (step S431).

In one round, I = in step S430.
Since it is set to 0, the code arrangement unit 10 proceeds to Yes from step S431, and then substitutes 0 into the variable J related to the channel number (step S432). Subsequently, the code arrangement unit 10 determines the variable J related to this channel number.
Is smaller than MAX_CH related to the maximum number of channels (step S433). Since J = 0 is set in step S432 in the first round, the code arrangement unit 10 proceeds to Yes from step S433 and arranges the differential bend value code [0] [0] (step S434). ),
Increment the variable J related to the channel number by 1 to J
= 1 (step S435).

In this second embodiment, since MAX_CH relating to the maximum number of channels is preset to 4, the code arrangement unit 10 thereafter repeats the processing of steps S433 to S435 to repeat the differential bend value code. [0] [1], differential bend value code [0] [2], differential bend value code [0] [3], and differential bend value code [0] [4] are sequentially arranged.

After arranging the differential bend value code [0] [4], the code arranging section 10 returns No from step S433.
After incrementing the variable I relating to the track number by 1, the processes of steps S431 to S435 are repeated again, and then the differential bend value code [1] [1] and the differential bend value code [1] [2] , The differential bend value code [1] [3], and the differential bend value code [1] [4] are sequentially arranged.

Also in this example, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG. Therefore, the code arrangement unit 10 performs the above step S4.
The processing from 31 to S435 is repeated, and finally the differential bend value code [m] [1], the differential bend value code [m] [2], the differential bend value code [m] [3], the differential relating to the track m data. When the bend value codes [m] [4] are sequentially arranged, this differential bend value code arrangement processing is ended, and the process returns to step S385 in FIG.

Subsequently, the code arrangement unit 10 arranges the difference volume value code (step S386). That is, FIG.
As shown in detail in the flowchart of FIG. 6, the code arrangement unit 10 first substitutes 0 into the variable I relating to the track number (step S440), and then determines whether or not this variable I is smaller than the number M of tracks. (Step S441).

[0331] In the first round, I = I in step S440.
Since the value is set to 0, the code arrangement unit 10 proceeds to Yes from step S441, and then substitutes 0 into the variable J related to the channel number (step S442). Subsequently, the code arrangement unit 10 determines the variable J related to this channel number.
Is smaller than MAX_CH related to the maximum number of channels (step S443). Since J = 0 is set in step S442 in the first round, the code arrangement unit 10 proceeds to Yes from step S443 and arranges the differential volume value code [0] [0] (step S44).
4), J related to the channel number is incremented by 1 to set J = 1 (step S445).

In the second embodiment, since MAX_CH relating to the maximum number of channels is preset to 4, the code arrangement unit 10 thereafter repeats the processing of steps S443 to S445 to repeat the differential volume value code. [0] [1], differential volume value code [0] [2], differential volume value code [0] [3],
Differential volume value codes [0] [4] are sequentially arranged.

After arranging the differential volume value code [0] [4], the code arranging unit 10 proceeds to the No side from step S443, increments the variable I relating to the track number by one, and then repeats the step. The processes of S441 to S445 are repeated, and then the differential volume value code [1] [1],
Differential volume value code [1] [2], differential volume value code
[1] [3] and differential volume value codes [1] [4] are sequentially arranged. In this example as well, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG.
Therefore, the code arrangement unit 10 repeats the processes of steps S441 to S445, and finally, the differential volume value code [m] [1], the differential volume value code [m] [2], the differential volume relating to the track m data. When the value code [m] [3] and the difference volume value code [m] [4] are sequentially arranged, the difference volume value code arrangement processing is ended, and the process returns to step S386 in FIG.

Next, the code arrangement unit 10 arranges the event type identification code (step S387). That is, FIG.
As shown in detail in the flowchart of FIG. 7, the code arrangement unit 10 first substitutes 0 into the variable I relating to the track number (step S450), and then determines whether or not this variable I is smaller than the number M of tracks. (Step S451).

In the first round, I = I in step S450.
Since it is set to 0, the code arrangement unit 10 proceeds from step S451 to Yes, and then substitutes 0 into the variable J related to the channel number (step S452). continue,
The code arrangement unit 10 determines whether or not the variable J related to this channel number is smaller than MAX_CH related to the maximum number of channels (step S453). In one round, the above step S
Since J = 0 is set in 452, the code arrangement unit 10
Advances from Step S453 to the Yes side, and after arranging the event type identification codes [0] [0] (Step S454),
The channel number J is incremented by 1 and J = 1
(Step S455).

In the second embodiment, since MAX_CH relating to the maximum number of channels is preset to 4, the code arrangement unit 10 thereafter repeats the processing of steps S453 to S455 to repeat the event type identification code. [0] [1], event type identification code [0] [2], event type identification code [0] [3],
Event type identification codes [0] [4] are sequentially arranged.

After the event type identification code [0] [4] is placed, the code placement unit 10 proceeds to step S4.
After No in 53, the variable I relating to the track number is incremented by 1, and then the steps S451 to S45 are performed again.
The process of 5 is repeated, and then the event type identification code [1]
[1], event type identification code [1] [2], event type identification code [1] [3], event type identification code [1] [4] will be sequentially arranged.

In the second embodiment, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG. Therefore, the code arrangement unit 10 repeats the processing of steps S451 to S455, and finally the event type identification code related to the track m data.
When the [m] [1], the event type identification code [m] [2], the event type identification code [m] [3], and the event type identification code [m] [4] are sequentially arranged, the event type identification code placement process is performed. Figure 5
0 returns to step S387.

The event type identification code is shown in FIG.
According to the event type identification code encoding table as shown in, it is appropriately assigned according to the event type.
It should be noted that it is possible to dynamically allocate according to the event type appearance frequency at the time of encoding, in which case it is more suitable for compression by Huffman encoding at the time of secondary encoding.

Subsequently, the code arrangement unit 10 arranges the channel number code (step S388).

That is, as shown in detail in the flowchart of FIG. 59, the code arrangement unit 10 first substitutes 0 into the variable I relating to the track number (step S460), and then this variable I is smaller than the number M of tracks. It is determined whether or not (step S461).

In one round, I = I in step S460.
Since it is set to 0, the code arrangement unit 10 proceeds from step S461 to the Yes side, arranges the channel number code [0] (step S462), and then increments the variable I related to the track number by 1 and I = 1 (step S463). Then, the code arrangement unit 10 repeats the processing of steps S461 to S463 again, and subsequently arranges the channel number code code [1]. In this example, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG. Therefore, the code arrangement unit 10
Repeats the processing of steps S461 to S463 and finally arranges the channel number code code [m] related to the track m data, terminates the channel number code arrangement processing, and returns to step S388 of FIG. Become.

Subsequently, the code arrangement unit 10 arranges other information codes (step S389).

That is, as shown in detail in the flowchart of FIG. 60, the code arrangement unit 10 first substitutes 0 into the variable I relating to the track number (step S470), and then this variable I is smaller than the number M of tracks. It is determined whether or not (step S471).

In one round, I = I in step S470.
Since it is set to 0, the code arrangement unit 10 proceeds from Step S471 to the Yes side, and then substitutes 0 for J related to the channel number (Step S472). Subsequently, the code arrangement unit 10 determines whether or not the variable J related to this channel number is smaller than MAX_CH related to the maximum number of channels (step S473). In one round, the above step S4
Since J = 0 is set in 72, the code arrangement unit 10
Advances from Step S473 to Yes, arranges the other information code [0] [0] (Step S474), and increments J related to the channel number by 1 to set J = 1 (Step S475). In the second embodiment, 4 is preset in MAX_CH related to the maximum number of channels.

Therefore, the code arranging unit 10 thereafter repeats the processing of steps S473 to S475 to repeat the other information code [0] [1], the other information code [0] [2], and the other information code [0]. [3] and other information codes [0] [4] will be sequentially arranged.

After arranging the other information code [0] [4], the code arranging section 10 proceeds to the No side from step S473, increments the variable I relating to the track number by 1 and sets I = 1. And then steps S471 to S4 again.
The process of 75 is repeated, and then the other information code [1]
[1], other information code [1] [2], other information code [1] [3],
Other information codes [1] [4] will be sequentially arranged.

In this example, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG. Therefore, the code arrangement unit 10 performs the above step S4.
The processing from 71 to S475 is repeated, and finally the other information code [m] [1] related to the track m data and the other information code
[m] [2], other information code [m] [3], other information code [m] [4]
Sequentially, the other information code allocation processing is terminated,
In this way, the code arrangement processing by the code arrangement unit 10 is completed.

Through the above processing, the codes are arranged in the basic arrangement order as shown in FIG. 61, and the primary code 5 is generated. That is, in the example of FIG. 61, more specifically, the basic arrangement order of the primary code 5 is as follows: Δ time code, pitch code, gate time code, velocity value code, differential bend value code, differential volume value code, event. The order is the type identification code, the channel number code, and the other information code. Further, since the internal arrangement of each code is the same as the content shown in FIG. 25, the duplicate description will be omitted here.

This primary code 5 is input to the secondary code generator 3 and compressed by the general-purpose compression method. In the second embodiment, the LZ encoding method, the Huffman encoding method, the run length encoding method, and a method combining these (for example, LHA or ZIP) may be adopted as the general-purpose compression technique. it can.

More specifically, assuming that the LZ encoding method is adopted, first, the processing is started from the beginning of the primary code 5, and the data pattern of the processing target position and the previously processed predetermined range are set. Compare with the data pattern of. When the two match, the information such as the distance from the processing target position to the data pattern and the length of the matched data pattern is output as the secondary code 6. On the contrary,
If they do not match, the primary code 5 is output as it is as the secondary code 6.

The above series of processing is repeated from the beginning to the end of the primary code 5.

In the second embodiment, as described above, since the primary code 5 in which each information is arranged in a temporally independent area for each information type and each channel is generated, the data pattern is repeated. This is also suitable for the LZ encoding method in which the repeated portion is compressed by utilizing this fact, and the compression rate can be lowered. Further, for example, with regard to the event type identification code, since it is possible to dynamically assign the code according to the frequency of occurrence of the event type, weighting is performed according to the frequency of occurrence of the code for each byte, and a small number of bits are assigned to a code having a weight. It is also suitable for the Huffman coding method for coding so that it can be recorded as a number, and the compression rate can be lowered. In addition to this, the velocity information is SMF.
Since there is often a bias in the distribution of values under the improved format of (1), recording the velocity information independently and collectively has a favorable effect on the efficiency of encoding by the Huffman encoding method.

In the second embodiment, it is assumed that the LZ method and the Huffman coding composite method (LHA) are adopted, but in that case, both effects described above are exhibited. Of course.

(Performance Information Decoding Device) Next, with reference to FIGS. 62 to 74, the performance information decoding device and the performance information decoding method according to the second embodiment for decoding the primary code or the secondary code. The performance information decoding program will be described in detail. The operation of the performance information decoding device corresponds to the performance information decoding method and the performance information decoding program according to the first embodiment.

The configuration of the performance information decoding apparatus 80 according to the second embodiment is substantially the same as the configuration shown in FIG. 26, and includes a secondary code decoding unit 2 and a primary code decoding unit 52. .

The secondary code decoding section 51 converts the secondary code 6 compressed by the general-purpose compression method described above into the primary code 5
To decrypt.

Further, the primary code decoding section 52 decodes the primary code 5 decoded by the secondary code decoding section 51 into the music file 4 in the above-mentioned format. In addition, this 1
The detailed configuration of the next code decoding unit 51 is substantially the same as the configuration shown in FIG. 27 described above, and the code extracting unit for each channel 110 (described later in FIG. 62), the track decoding unit 120, the track merging and the channel number code decoding. 55, track placement section 56
have.

The structure and operation of each of these parts will be described in detail below.

First, the detailed configuration of the code-by-channel extraction section 110 is as shown in FIG. 62, and each code included in the primary code 5 is extracted for each channel.

That is, the channel code extraction section 110
From the primary code 5, track 0, channel 0 code group, track 0, channel 1 code group, ..., Track m, channel n code group are sequentially extracted for each channel. Each of these code groups includes a Δ time code, a pitch code, a gate time code, a velocity value code, a differential bend value code, a differential volume value code, an event type identification code, and other information codes.

The flow of the processing relating to the extraction of the channel code by the channel code extraction section 110 will be described in detail below with reference to the flowchart in FIG.

Now, when entering this processing, the channel code extraction section 110 first extracts the code of the header chunk (step S480). After that, each code is extracted.

First, the channel code extraction unit 110 extracts the Δtime code (step S481). That is, FIG.
As shown in detail in the flowchart of FIG. 4, the channel-specific code extraction unit 110 first determines a variable I relating to the track number.
After substituting 0 for (step S490), it is determined whether or not this variable I is smaller than the number M of tracks (step S491). In one round, I =
Since it is set to 0, the channel-specific code extraction unit 110
Proceeds from Step S491 to Yes, and then assigns 0 to the variable J related to the channel number (Step S49
2). Next, the channel-specific code extraction unit 110 determines that the variable J related to this channel number is MAX_CH related to the maximum number of channels.
It is determined whether it is smaller than (step S493).

In the 1st round, J = in step S492.
Since it is set to 0, the channel-specific code extraction unit 110
Advances from Step S173 to the Yes side, extracts the Δtime code [0] [0] (Step S494), and increments the variable J related to the channel number by 1 to set J = 1 (Step S495).

In this second embodiment, since MAX_CH relating to the maximum number of channels is preset to 4, the code-by-channel code extraction unit 110 will thereafter perform the above step S4.
The processing from 93 to S495 is repeated, and the Δ time code [0]
[1], Δ time code [0] [2], Δ time code [0] [3], and Δ time code [0] [4] are sequentially extracted.

After extracting the Δ time code [0] [4], the channel-specific code extraction section 110 advances from step S493 to No, increments the variable I relating to the track number by one, and then repeats the above. Steps S491 to S49
The processing of 5 is repeated, and the Δtime code [1] [1], the Δtime code [1] [2], the Δtime code [1] [3], and the Δtime code [1] [4] are sequentially extracted. In this example as well, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG. Therefore, the channel-specific code extraction unit 110 repeats the processing of steps S491 to S495,
Finally, Δ time code [m] [1] related to track m data,
When the Δtime code [m] [2], the Δtime code [m] [3], and the Δtime code [m] [4] are sequentially extracted, this delta time code extraction processing is ended, and step S481 of FIG. 63 is performed. Return to.

Subsequently, the channel code extraction unit 110 extracts the pitch code (step S482). That is, FIG.
As shown in detail in the flowchart of FIG. 3, the channel-specific code extraction unit 110 first substitutes 0 into the variable I relating to the track number (step S500), and then determines whether this variable I is smaller than the number M of tracks. Judge (step S
501). For one round, I = 0 in step S500.
Is set, the channel-specific code extraction unit 110
Proceeds from Step S501 to Yes, and then substitutes 0 for J relating to the channel number (Step S50
2). Next, the channel-specific code extraction unit 110 determines that the variable J related to this channel number is MAX_CH related to the maximum number of channels.
It is determined whether it is smaller than (step S503).

[0369] In the first round, J = J in step S502.
Since it is set to 0, the channel-specific code extraction unit 110
Goes to the Yes side from step S503, and the pitch code
[0] After extracting [0] (step S504), the variable J related to the channel number is incremented by 1 to set J = 1 (step S505).

In this second embodiment, since MAX_CH related to the maximum number of channels is preset to 4, the code-by-channel code extraction unit 110 thereafter repeats the processing of steps S503 to S505 to perform the pitch code [ 0] [1], pitch code
[0] [2], pitch code [0] [3], and pitch code [0] [4] are sequentially extracted. The channel code extraction unit 110 uses the pitch code [0].
After extracting [4], the process proceeds to the No side from step S503, increments the variable I relating to the track number by one, and then repeats the processes of steps S501 to S505, and then the pitch code [1] [1] The pitch code [1] [2], the pitch code [1] [3], and the pitch code [1] [4] are sequentially extracted.
In this example as well, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG.

Therefore, the channel code extraction section 110
Repeat the above steps S501 to S505,
Finally, when the pitch code [m] [1], the pitch code [m] [2], the pitch code [m] [3], and the pitch code [m] [4] related to the track m data are sequentially extracted, this pitch The code extraction process is terminated, and the process returns to step S482 in FIG.

Subsequently, the channel code extraction unit 110 extracts the gate time code (step S483). That is, as shown in detail in the flowchart of FIG.
The channel-specific code extraction unit 110 first substitutes 0 into the variable I relating to the track number (step S510), and then determines whether or not this variable I is smaller than the number M of tracks (step S511). First, in the first round, since I = 0 is set in step S510, the channel-specific code extraction unit 110 proceeds from step S511 to the Yes side, and then sets variable J related to the channel number to 0 ( Step S512). Subsequently, the channel code extraction unit 110 determines whether the variable J related to this channel number is smaller than MAX_CH related to the maximum number of channels (step S513).

For one round, J = J in step S512.
Since it is set to 0, the channel-specific code extraction unit 110
Advances from step S513 to Yes, extracts the gate time code [0] [0] (step S514), and increments the variable J related to the channel number by 1 to set J = 1 (step S515). ).

In this second embodiment, since MAX_CH related to the maximum number of channels is preset to 4, the channel-by-channel code extraction unit 110 thereafter repeats the processing of steps S513 to S515 to repeat the gate time code. [0] [1],
The gate time code [0] [2], the gate time code [0] [2], and the gate time code [0] [4] are sequentially extracted.

Then, the code extraction unit for each channel 110
After the gate time code [0] [4] is extracted, the process proceeds to the No side from step S513, and the variable I relating to the track number.
Is incremented by 1 to set I = 1, and then the processes of steps S511 to S515 are repeated again, and subsequently, the gate time code [1] [1], the gate time code [1] [2], and the gate time code [1 ] [3] and gate time codes [1] [4] are sequentially extracted.

Also in this example, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG. Therefore, the channel-specific code extraction unit 110 repeats the processing of steps S511 to S515, and finally the gate time code [m] related to the track m data.
[1], Gate time code [m] [2], Gate time code [m]
[3] and the gate time code [m] [4] are extracted sequentially, this N
The ote Off pitch code extraction process is terminated, and the process returns to step S483 in FIG.

Subsequently, the channel code extraction unit 110 extracts the velocity value code (step S484). That is, as shown in detail in the flowchart of FIG.
The channel-specific code extraction unit 110 first substitutes 0 into the variable I relating to the track number (step S520), and then determines whether this variable I is smaller than the number M of tracks (step S521). In one round, the above step S5
Since I = 0 is set at 20, the channel-by-channel code extraction unit 110 proceeds to Yes from step S521, and then substitutes 0 for the variable J related to the channel number (step S522). Then, the channel-specific code extraction unit 110
Determines whether the variable J related to this channel number is smaller than MAX_CH related to the maximum number of channels (step S
523).

[0378] In the first round, J = J in step S522.
Since it is set to 0, the channel-specific code extraction unit 110
Advances from step S523 to Yes, extracts velocity value code [0] [0] (step S524), and increments J related to the channel number by 1 to set J = 1 (step S525). .

In this second embodiment, 4 is preset in MAX_CH related to the maximum number of channels. Therefore, the channel-by-channel code extraction unit 110, after that, in step S
523 to S525 are repeated, and the velocity value code [0] [1], velocity value code [0] [2], velocity value code
[0] [3] and velocity value codes [0] [4] are sequentially extracted.

Then, the code extraction unit for each channel 110
After extracting the velocity value code [0] [4], the process proceeds to No side from step S523, and the variable I relating to the track number
Is incremented by 1 to set I = 1, and then the processes of steps S521 to S525 are repeated, and subsequently, the velocity value code [1] [1], the velocity value code [1] [2], and the velocity value code [1 ] [3] and velocity value codes [1] [4] are sequentially extracted.

Also in this example, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG. Therefore, the channel-specific code extraction unit 110 repeats the processes of steps S521 to S525, and finally the velocity value code [m] related to the track m data.
[1], velocity value sign [m] [2], velocity value sign [m]
When [3] and velocity value code [m] [4] are sequentially extracted, this velocity value code extraction processing is ended, and the process returns to step S484 in FIG.

Subsequently, the channel code extraction unit 110 extracts the differential bend value code (step S485). That is, as shown in detail in the flowchart of FIG.
The channel-specific code extraction unit 110 first sets 0 to the variable I relating to the track number (step S530), and then determines whether this variable I is smaller than the number M of tracks (step S531). In one round, the above step S5
Since I = 0 is set in 30, the channel-specific code extraction unit 110 advances to Yes from step S531,
Next, 0 is substituted for J related to the channel number (step S532). Then, the channel code extraction unit 110
The variable J related to this channel number relates to the maximum number of channels
It is determined whether it is smaller than MAX_CH (step S53).
3).

In one round, J = in step S532.
Since it is set to 0, the channel-specific code extraction unit 110
Proceeds from Step S533 to the Yes side, and after extracting the difference bend value code [0] [0] (Step S534), the variable J related to the channel number is incremented by 1 and J = 1.
(Step S535).

In this second embodiment, since MAX_CH related to the maximum number of channels is preset to 4, the code-by-channel code extraction unit 110 thereafter repeats the processing of steps S533 to S535 to obtain the differential bend value. Code [0] [1],
The differential bend value code [0] [2], the differential bend value code [0] [3], and the differential bend value code [0] [4] are sequentially extracted.

Then, the code extraction unit for each channel 110
After extracting the difference bend value code [0] [4], the process proceeds from the step S533 to the No side, and the variable I relating to the track number.
After incrementing by one, steps S531 to S531
The processing of 535 is repeated, and then the differential bend value code [1]
[1], differential bend value code [1] [2], differential bend value code [1]
[3] and differential bend value codes [1] [4] will be sequentially extracted.

Also in this example, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG. Therefore, the channel-specific code extraction unit 110 repeats the processing of steps S531 to S535, and finally the differential bend value code [m] related to the track m data.
[1], differential bend value code [m] [2], differential bend value code [m]
[3] and the differential bend value code [m] [4] are sequentially extracted, the differential bend value code extraction processing ends, and the process returns to step S485 in FIG.

Subsequently, the channel code extraction section 110
The differential volume value code is extracted (step S48).
6). That is, as shown in detail in the flowchart of FIG. 69, the channel-specific code extraction unit 110 first substitutes 0 into the variable I relating to the track number (step S54).
0), it is determined whether or not this I is smaller than the number M of tracks (step S541). In the first round, since I = 0 is set in step S540, the channel-specific code extraction unit 110 advances to Yes from step S541, and then substitutes 0 into the variable J related to the channel number (step S542). ). Next, the channel code extraction unit 110 determines whether the variable J related to this channel number is smaller than MAX_CH related to the maximum number of channels (step S543).

In this first round, since J = 0 is set in step S542, the code extraction unit for each channel 1
In step 10, the process proceeds from step S543 to the Yes side, and after extracting the differential volume value code [0] [0] (step S54
4), J related to the channel number is incremented by 1 to set J = 1 (step S545). In the second embodiment, 4 is set in advance in MAX_CH related to the maximum number of channels.

Therefore, the channel code extraction section 110
After that, the processes of steps S543 to S545 are repeated, and the difference volume value code [0] [1], the difference volume value code [0] [2], the difference volume value code [0] [3], the difference volume value code [ 0] [4] are sequentially extracted. After extracting this differential volume value code [0] [4], the channel-specific code extraction unit 110 proceeds to No side from step S543, increments the variable I related to the track number by one, and then repeats steps S541 to S541. The process of S545 is repeated, and then the differential volume value code [1] [1] and the differential volume value code
[1] [2], differential volume value code [1] [3], differential volume value code [1] [4] are sequentially extracted.

Also in this example, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG. Therefore, the channel-specific code extraction unit 110 repeats the processing of steps S541 to S545, and finally the differential volume value code [m] related to the track m data.
When the differential volume value code [m] [2], differential volume value code [m] [3], and differential volume value code [m] [4] are sequentially extracted, the differential volume value code extraction processing ends. , FIG. 63
Return to step S486.

Next, the channel code extraction section 110 extracts the event type identification code (step S487).
That is, as shown in detail in the flowchart of FIG. 70, the channel-specific code extraction unit 110 first substitutes 0 into the variable I relating to the track number (step S550),
It is determined whether or not this variable I is smaller than the number M of tracks (step S551). First, in one round, since I = 0 is set in step S550, the channel-specific code extraction unit 110 proceeds to Yes from step S551, and then sets 0 to the variable J related to the channel number (step S551). S552). Subsequently, the channel code extraction unit 110 determines whether the variable J related to this channel number is smaller than MAX_CH related to the maximum number of channels (step S553).

In the first round, J = J in step S552.
Since it is set to 0, the channel-specific code extraction unit 110
Proceeds from Step S553 to the Yes side, and after extracting the event type identification code [0] [0] (Step S554),
The channel number J is incremented by 1 and J = 1
(Step S555).

In the second embodiment, since MAX_CH relating to the maximum number of channels is preset to 4, the channel-specific code extraction unit 110 thereafter repeats the processing of steps S553 to S555 to determine the event type. Identification code
[0] [1], event type identification code [0] [2], event type identification code [0] [3], and event type identification code [0] [4] are sequentially extracted. After extracting the event type identification code [0] [4], the channel code extraction unit 110 proceeds to No side from step S553, increments I related to the track number by 1, and then again steps S551 to S555. The above process is repeated, and then event type identification code [1] [1], event type identification code [1] [2], event type identification code [1]
[3] and event type identification codes [1] and [4] are sequentially extracted.

In this example, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG. Therefore, the channel-specific code extraction unit 110 repeats the processing of steps S551 to S555, and finally the event type identification code [m] related to the track m data.
When [1], the event type identification code [m] [2], the event type identification code [m] [3], and the event type identification code [m] [4] are sequentially extracted, the event type identification code extraction process ends. , FIG. 63
Return to step S487.

Subsequently, the channel code extraction section 110
The channel number code is extracted (step S488). That is, as shown in detail in the flowchart of FIG.
The channel-specific code extraction unit 110 first substitutes 0 into the variable I relating to the track number (step S560), and then determines whether or not this variable I is smaller than the number M of tracks (step S561). First, in the first round, since I = 0 is set in step S560, the channel-specific code extraction unit 110 proceeds to Yes from step S561 and extracts the channel number code [0] (step S561).
562), the variable I relating to the track number is incremented by 1 to set I = 1 (step S563).

Then, the code extraction unit for each channel 110
The processing of steps S561 to S563 is repeated again,
Then, the channel number code code [1] is extracted.

In this example, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG. Therefore, the channel-specific code extraction unit 110 repeats the processing of steps S561 to S563, and finally the channel number code code [m] relating to the track m data.
When is extracted, the channel number code extraction processing is ended, and the process returns to step S488 in FIG.

Subsequently, the channel code extraction section 110
Other information codes are extracted (step S489). That is, as shown in detail in the flowchart of FIG.
The channel-specific code extraction unit 110 first substitutes 0 into the variable I relating to the track number (step S570), and then determines whether this I is smaller than the number M of tracks (step S571). First, in one round, the above step S5
Since I = 0 is set at 70, the channel-specific code extraction unit 110 proceeds to Yes from step S571, and then substitutes 0 for the variable J related to the channel number (step S572).

Subsequently, the channel code extraction section 110
The variable J related to this channel number is the maximum number of channels MAX_CH
It is determined whether it is smaller than (step S573).

In this first round, since J = 0 is set in step S572, the code extraction unit for each channel 1
In step 10, the process proceeds from step S573 to the Yes side, and after extracting the other information code [0] [0] (step S574), the variable J related to the channel number is incremented by 1 and J =
1 (step S575).

In the second embodiment, since MAX_CH related to the maximum number of channels is preset to 4, the channel-by-channel code extraction unit 110 thereafter repeats the processing of steps S573 to S575 to obtain other information. Code [0] [1], other information code [0] [2], other information code [0] [3], and other information code [0] [4] are sequentially arranged.

After extracting the other information code [0] [4], the channel code extraction section 110 returns No to step S573.
After incrementing the variable I relating to the track number by 1, the process of steps S571 to S575 is repeated again, and subsequently, other information code [1] [1], other information code [1] [2], etc. Information code [1] [3], other information code
[1] [4] are sequentially extracted. In this example, it is assumed that m + 1 related to the number of tracks is set to m + 1 as shown in FIG.

Therefore, the channel code extraction unit 110
By repeating the processes of steps S251 to S255,
Finally, other information code [m] related to track m data
[1], other information code [m] [2], ..., other information code [m] [1
5] is sequentially extracted, the other information code extraction process is ended, and thus, a series of processes related to the code extraction for each channel by the code extraction unit for each channel 110 is ended.

That is, by the above processing, a code group consisting of a single channel (each code group is a Δ time code, a pitch code,
The gate time code, the velocity value code, the difference bend value code, the difference volume value code, the event type identification code, and the other information code are included, of course).

Next, the detailed structure of the track decoding unit 120 is as shown in FIG.
21, pitch code decoding unit 122, gate time code decoding unit 123, velocity value code decoding unit 124, differential bend value code decoding unit 125, differential volume value code decoding unit 12
6, an event type identification code decoding unit 127, an other information code decoding unit 128, and an event writing unit 129.

Under such a configuration, each of these units 121 to 128 decodes each code, and the event writing unit 129 writes an event for each code, thus making a track consisting of a single channel. Generate data.

[0407] The flow of the process relating to the track decoding by the track decoding unit 54 is substantially the same as that shown in Fig. 40 described above, and Track performed at step S265 of Fig. 40.
The processing relating to the decoding of [I] [J] is also substantially the same as in FIG. However, the process related to decryption for each event type is different from that in FIG. 42.

Therefore, hereinafter, with reference to the flowchart of FIG. 74, the process relating to the decoding for each event type according to the second embodiment will be described in detail.

[0409] That is, when this process is started, the track decoding unit 54 determines the type of event from the Δtime (step S600). In the second embodiment, Δ
The type of event can be identified by the information in the first byte of the time.

In this example, the event types can be identified as three types (type 1 to type 3) of events. Therefore, the track decoding unit 54 determines the type of event based on the information of the first byte of the Δtime.

[0411] In step S600, when the track decoding unit 54 determines that the event is type 1, the track decoding unit 54 reads 1-byte information from the pitch code [I] [J], and the upper 2 bits are J and the lower 2 bits. Assign the key number to 6 bits and append to the end of Event (step S60)
1). At this time, the key shift is saved.

Next, the track decoding section 54 reads 1-byte information from the gate time code [I] [J],
Is appended to the end (step S602). further,
The track decoding unit 54 determines 1 from the velocity value code [I] [J].
Reads the byte information, the upper 6 bits are velocity,
The key shift is assigned to the lower 2 bits, the event is appended to the end of the event (step S603), and the process proceeds to step S620.

When it is determined in step S600 that the event is of type 2, the track decoding unit 54 generates a status byte 0xff and appends it to the end of the event (step S604). Then, the track decoding unit 54 generates a command byte in which 0 is assigned to the upper 6 bits and J is assigned to the lower 2 bits, and the command byte is appended to the end of the Event (step S605). Further, the track decoding unit 54 reads 1-byte information from the pitch code [I] [J] and appends it to the end of Event (step S60).
6) Read 1-byte information from the gate time code [I] [J] and append it to the end of Event (step S60
7) and proceeds to step S620.

On the other hand, in step S600, when the track decoding section 54 determines that the event is type 3, the track decoding part 54 sequentially reads one code from the event type identification codes [I] [J] (step S608), It is determined whether the event type identification code is 0x08 or more (step S609). Here, if the event type identification code is 0x08 or more, the track decoding unit 54 generates a status byte 0xff, appends it to the end of Event (step S610), and generates a command byte corresponding to the event type identification code. And append to the end of Event (step S
611), and it is determined whether the event type identification code is 0x0d (step S612).

If the event type identification code is 0x0d in step S612, the track decoding unit 54 reads 1 byte from the differential bend value code [I] [J], and adds the value added to the immediately preceding bend value. Substitute 6 for the lower 6 bits and J for the upper 2 bits, and append to the end of the Event (step S6
13) and proceeds to step S620.

On the other hand, if the event type identification code is not 0x0d in step S612, the track decoding unit 54 subsequently determines whether the event type identification code is 0x0b (step S614), and it is 0x0b. Then, 1 byte is read from the differential volume value code [I] [J], and the value added to the immediately preceding volume value is set to the lower 6 bits,
Substitute J for the upper 2 bits, append to the end of Event (step S615), and proceed to step S620. On the other hand, if the event type identification code is not 0x0b in step S614, the track decoding unit 54
Then reads 1-byte information from the other information code [I] [J], substitutes 0 for the lower 6 bits and J for the upper 2 bits, and appends to the end of the Event (step S616).
The process proceeds to step S620. Step S6 above
In 09, if the event type identification code is less than 0x08, the track decoding unit 54 generates a status byte according to J, appends it to the end of Event (step S617), and the event type identification information is 0x03 or more. It is determined whether or not there is (step S618).

Here, J indicates a channel number, but as described above, when encoding the improved format of SMF, depending on the event class, channel 0 is for Class Normal, channel 1 for Class A, and channel for Class B. If 2, Class C is handled as an event belonging to channel 3, in step S617, the status byte corresponding to the event class is restored and generated.

If the event type identification code is 0x03 or more in step S618, the track decoding unit 54 generates a command byte corresponding to the event type identification code and appends it to the end of the event (step S61).
9) and proceeds to step S620. On the other hand, if the event type identification code is less than 0x03 in step S618, the track decoding unit 54 proceeds directly to step S620. Thus, in step S620, the track decoding unit 54 follows the specifications of the file described above.
The code is read from the other information code [I] [J] until the event is completed, and the code is appended to the end of the event (step S6).
20) and this process is completed.

Next, the detailed structure of the track merge / channel number code decoding unit 55 is substantially the same as the structure shown in FIG. 43, and includes a track merge unit 55a and a channel number code decoding unit 55b. Under such a configuration, the channel number code decoding unit 55b decodes the channel number code 13, and the track merge unit 55a decodes the track x data 11-x while adding the channel number code. At this time, as for the event that does not belong to the channel, as described in the description of encoding, it is assumed that the event belongs to the channels 0 to 3 corresponding to the event class, and the channel number code 1
Since it is recorded in No. 3, it will be decrypted by taking this correspondence. The flow of the processing relating to the decoding by the track merging unit 55 is as previously shown in the flowcharts of FIGS. 44 and 45, and therefore, the duplicate description will be omitted here.

Thus, when track 0 data, track 1 data, ..., Track m data for each track is generated, these are input to the track arranging section 56 and arranged in the music file 4 by the track arranging section 56. To be done.

The details of the processing relating to the track arrangement by the track arrangement unit 56 are the same as those shown in the flow chart of FIG. 47, and the duplicated description will be omitted.

By the series of processing described above, 2
The secondary code 6 to the secondary code decoding unit 51 decodes the primary code 5, and the primary code 5 to the primary code decoding unit 52 decodes the music file 4.

As described above, the performance information compression apparatus, the performance information compression method, the performance information compression program, the performance information decoding apparatus, the performance information decoding method, and the performance information decoding program according to the second embodiment of the present invention are The following effects are exhibited.

That is, in the performance information compression device and the like, the processing by the primary code generation unit 2 eliminates waste of the music file 4 (improvement of SMF), and furthermore, the music file 4 (SMF).
It is realized that the primary code 5 having the compressed data amount is generated without degrading the performance information in (1.
Further, the primary code 5 has a higher appearance frequency of the same data pattern than the music file 4 (improved SMF), the same data pattern is recorded at a closer distance, and the same data is recorded in the code of the same type. Since the probability that the pattern appears is high, the compression efficiency by the general-purpose compression method in the secondary code generation unit 3 is improved.

This effect is similarly exhibited in the performance information decoding device and the like in the primary code decoding unit 52 and the secondary code decoding unit 51. Moreover, by preserving the information relating to the order of each performance information while temporally making each performance information independent at the time of the primary encoding, the order of each performance information is decoded when the encoded performance information is decoded. Does not change. That is, the decrypted performance information completely matches the original performance information.

(Third Embodiment) In the third embodiment, the performance information compression apparatus, the performance information compression method, the performance information compression program, and the performance information decoding according to the first and second embodiments described above. An example in which the device, the performance information decoding method, and the performance information decoding program are applied to a system in which a mobile terminal device downloads an incoming melody will be shown and described.

In the third embodiment, the format of the input data can be applied not only to the SMF and its improved format but also to the ringing melody dedicated format such as MFi, SMAF, and compact MIDI.

First, FIG. 75 shows a first structural example of a system according to the third embodiment.

[0429] The SMF 200 and the lyrics file 201 as performance information are input to the conversion section 202 and converted into a performance information file 203. In this lyrics file 201,
It describes the text information of the lyric character, the timing information of the color change of the text, the display position and typeface of the text, the page switching timing and the like. Even if the performance information file 203 is a dedicated file format capable of describing pitch information, sound intensity information, sound length information, and other information It may be a melody-dedicated format or SMF. The performance information file 203 is primary-coded by the primary code generation unit 204, and then secondary-coded by the secondary code generation unit 206 to generate a compressed file 208.

At the time of this secondary encoding, the reproduction player (software) 205 is also included in the compressed file 208. As the primary code generation unit 204, the primary code generation unit 2 according to the first and second embodiments can be adopted, and as the secondary code generation unit 206, the primary code generation unit 206 of the first and second embodiments can be used. The secondary code generation unit 3 according to the mode can be adopted. The compressed file 208 is uploaded to the server 207.

The compressed file 208 is downloaded to the mobile terminal device 300 based on the request from the mobile terminal device 300. The downloaded compressed file 208 is stored in the storage area 209 of the mobile terminal device 300.

The compressed file 208 is sent to the secondary code decoding unit 210, is decoded in the secondary code decoding unit 210, and is separated into the reproduction player 212 and the primary code 208. The primary code 208 is stored in the storage area 211.

The playback player 212 includes a primary code decoding section 213 and a lyrics display section 214, and the primary code 208 is decoded by the primary code decoding section 213 to obtain a performance information file 203. , Are stored in the storage area 215. This performance information file 203 is SMF2
00 and lyrics files etc. 201 are separated, SMF200
Is sent to the voice reproduction unit 216, and the lyrics file 201 is sent to the lyrics display unit 214. Then, the audio reproduction unit 21
6, the voice is reproduced, and the lyrics display section 214 displays the lyrics. Since the voice reproducing unit 216 and the lyrics display unit 214 can adopt a known method, detailed description thereof will be omitted.

Next, FIG. 76 shows a second configuration example of the system according to the third embodiment.

Also in this second configuration example, the SMF 200
And the lyrics file 210 are input to the conversion unit 202,
The performance information file 203 is generated in the conversion unit 202 as in the first configuration example. However, in the second configuration example, the performance information file 203 is compressed by the first compression unit 220, and the compressed file 222 is generated. As the first compression unit 220, the performance information compression device according to the above-described first and second embodiments can be adopted. On the other hand, the reproduction player 205 for reproducing the performance information file 203 includes a first decoding unit 225 for decoding the performance information file 222. The reproduction player 205 is compressed by the second compression unit 221 to generate the compressed file 223. These compressed files 222 and 223 are uploaded to the server 207.

Both of these compressed files 222 and 223 are downloaded to the storage area 209 of the portable terminal device 301. Then, of the downloaded compressed files 222 and 223, the compressed file 223 relating to the reproducing player is decoded by the second decoding unit 224, and the reproducing player 225 is obtained.

On the other hand, the compressed file 222 relating to the performance information file 203 is decompressed by the first decoding section 226,
The performance information file 203 is generated.

Then, this performance information file 203 is
The SMF 200 and the lyrics file 201 are expanded, the lyrics file 201 is sent to the lyrics display unit 214, and the lyrics are displayed on the lyrics display unit 214.
00 is sent to the audio reproducing unit 216 and the audio reproducing unit 21
6, the sound is reproduced.

As the first decoding section 226, the performance information decoding apparatus according to the above-mentioned first and second embodiments can be adopted.

Next, FIG. 77 shows a third configuration example of the system according to the third embodiment.

Here, the difference from the first and second configuration examples described above will be mainly described.

In this third configuration example, the mobile terminal device 302
The reproduction player 228 in the inside has a second decoding unit 226, a lyrics display unit 214, and an audio reproduction unit 227 built therein.

As the first decoding section 226, it is a matter of course that the performance information decoding apparatus according to the above-mentioned first and second embodiments can be adopted. The first compression unit 220 uses a unique compression method that does not disclose the algorithm, and uses the first compression unit 220 and the first decoding unit 22.
By making 6 undisclosed, it is possible to prevent an illegal server due to illegal production of contents.

Next, FIG. 78 shows a fourth configuration example of the system according to the third embodiment.

Here, differences from the above-described first to third configuration examples will be mainly described.

In the second and third configuration examples, the SMF 20
0 and the lyrics file 201 are converted into the performance information file 203 by the conversion unit 202 and compressed by the first compression unit 220, whereas the lyrics file 201 is not compressed in the fourth configuration example. The performance information file 203 is combined with the compressed file 222 compressed by the first compression unit 220. And this lyrics file 2
The performance information file 222 in which 01 is combined is stored in the server 2
It will be uploaded to 07.

The performance information file 222 and the compressed file 223 relating to the reproduction player 205 are downloaded to the storage area 209 of the portable terminal device 303. When using the compressed files 222 and 223,
First, the compressed file 223 is decoded by the second decoding unit 224, and the reproduction player 228 is obtained.

On the other hand, the lyrics file 201 is separated from the performance information file 222 in which the lyrics file 102 is combined. Then, the compressed file 229 from which the lyrics file 201 has been separated is expanded by the first decoding unit 226, the performance information file 203 is decoded, and further SM
It is expanded to F200, stored in the storage area 215, and reproduced by the sound reproducing unit 216. On the other hand, the separated lyrics file 201 is sent to the lyrics display section 214, and the lyrics are displayed here.

As the first decoding section 226, the performance information decoding apparatus according to the above-mentioned first and second embodiments can be adopted.

In the first to fourth configuration examples described above,
In order to decode the compressed performance information file 208, the primary code decoding unit 213 attached to the reproduction player 106 is required, so the performance information file 203
If a process such as associating the reproduction information with the reproduction player 205 and adding an ID or digital watermark to the reproduction information file 203 and allowing the reproduction information to coincide with each other is performed, a compressed performance information file 208 is obtained. Since it cannot be used even if exchanged alone, there is an effect that illegal reproduction by different reproduction players is prevented.
The performance information file 203 is subjected to the primary coding and the secondary coding by the primary code generation unit 2 and the secondary code generation unit 3 of the performance information compression device described in the first and second embodiments. Since it is adopted, the amount of data can be further reduced.

Further, although the portable terminal device 301 is taken as an example in the above-mentioned first to fourth configuration examples, the present invention is not limited to this, and the memory capacity of home electric appliances, in-vehicle terminals, fixed telephone terminals, etc. It is applicable to any system that needs to limit the network traffic capacity. Further, in the case where a file having a limited capacity, for example, a file downloaded to a recording medium such as a memory card is stored as it is, the storage capacity of the memory card can be saved.

According to the third embodiment described above,
Since the data amount of the performance information file can be reduced, the time for downloading the file to the mobile terminal device can be shortened and the capacity of the storage means of the mobile terminal device can be saved.

If the reproduction player and the compressed performance information file are transmitted at the same time so that the performance information file cannot be reproduced by another reproduction player, unauthorized use of the performance information file can be prevented. Also, for example, the first and second 100-Kbyte SMFs described above are used.
When the device according to the embodiment of the present invention is applied to the system according to the third embodiment, it is possible to compress up to about 10 kilobytes, so that it is more suitable in terms of size when exchanging things such as ringing melody. It will be

Although the first to third embodiments of the present invention have been described above, in the performance information compression apparatus, method and program of the present invention, the SMF or its improved file is provided for each channel at least in the relative time between events. Information of
It is separated into the information of the starting pitch, the information of the strength of the sound, the information of the ending pitch of the sound, and other information (the length of the sound in the improved file), and for each of the above information regardless of the above channel. In summary, a primary code is generated by arranging the collected pieces of information in independent areas, and the information of each area of the primary code is compressed by a general-purpose compression method. SMF
However, one track chunk includes one track data, and the track data may include data related to a plurality of channels. However, as described above, by grouping each information regardless of the channel, A suitable primary code can be generated by the compression by the compression method, and further, the secondary code can be compressed with high efficiency.

Further, in the performance information compression apparatus, method and program of the present invention, when a plurality of events at the same time are recorded in the same track data at the time of generating the primary code, an event consisting of a plurality of events Information on channels other than the event recorded at the end of the group is recorded. In this way, the data amount can be reduced by deleting the channel information of the event recorded at the end.

The performance information compression apparatus, method, and
In the program, when the primary code is generated, different event type identification codes are assigned to each event depending on the frequency of appearance, and the event type identification code is recorded for each event. The LHA coding method that combines the LZ method and the Huffman coding method can also be used for the secondary code of the present invention, but if the event type identification code is assigned according to the frequency of occurrence, the dynamic Huffman coding method can be used. It becomes more preferable, and as a result, the compression rate can be lowered.

Then, the performance information compression apparatus, method and program of the present invention identify the event type based on the information of the relative time between events when the primary code is generated. According to this, for example, the event type can be identified based on the Δtime information, the data amount is reduced, and the compression efficiency at the time of secondary coding is further enhanced.

Furthermore, the performance information compression apparatus, method and program of the present invention calculate the greatest common divisor of the relative time between each event at the time of the primary code and divide the relative time between each event by the above greatest common divisor. Encode numbers. The Δtime exists as a pair with each event, but even in the case where it normally takes 2 bytes, it can be completed by using 1 byte, and the capacity can be reduced as a whole. If the greatest common denominator cannot be calculated, it may be possible to substitute a predetermined value.

Also, the performance information decoding apparatus, method, and
Since the program decodes the secondary code compressed by the performance information compression device or the like to obtain the primary code, and further decodes the primary code to obtain the music file, the same effect as described above is obtained. Played.

The present invention is not limited to the first to third embodiments described above, and various improvements and changes can be made without departing from the spirit of the invention.

For example, in the above embodiment, the LHA coding method, which is a decoding method of the LZ coding method and the Huffman coding method, is adopted at the time of the secondary coding.
Without being limited to this, various compression methods can be adopted. For example, run length coding method, BSTW
Such as a coding method and a ZIP coding method.

Regarding this LZ encoding method, LZ7
7 encoding method, as a variation, LZR
Encoding method, LZSS encoding method, LZB encoding method, LZH
There are various encoding methods such as an encoding method, an LZBW encoding method, and an LZIT encoding method.

Furthermore, the format of the input data (music file) is not limited to the formats described at the beginning of the first and second embodiments, but SMA
Of course, it is also applicable to F, MFi, and CMX formats.

The value of the maximum channel number MAX_CH in each flow chart is not limited to the above value.

The following technical ideas are included in the present invention.

First, the track data belonging to a predetermined channel, which includes at least an event consisting of pitch information, sound intensity information, and other information, and information of relative time between the events, is assigned. In a performance information transmission system that receives an input of a performance information file that it has and compresses and transmits the performance information file, the track data of the performance information file is separated into a plurality of track data belonging to each channel, and the track data is assigned to the channel. While holding the information indicating the attribution of each, the track data belonging to each channel is further separated for each of the above information, and the separated information is summarized for each type of information regardless of the attribution to the channel, It is generated by the primary code generation means for generating a primary code by arranging it in an independent area for each type, and the primary code generation means. Playback player program form for reproducing the information and the performance information files placed in the area of the next code is compressed by a predetermined compression method,
A performance information transmission system, comprising: a secondary code generation means for generating a secondary code; and a transmission means for transmitting the secondary code.

Secondly, track data belonging to a predetermined channel, which includes at least an event consisting of pitch information, sound intensity information, and other information, and information of relative time between the events, is assigned. A performance in which a performance information file that the player has and a playback player in a program form for reproducing the performance information file are compressed to decode a generated secondary code into a primary code, and further the primary code is decoded into a performance information file. In the information transmission system, transmitting means for receiving the secondary code, secondary code decoding means for decoding the secondary code into a primary code and a playback player in a program form for playing the performance information file, and Regarding the primary code, each type of information arranged in an independent area is extracted for each type, and the information indicating attribution to the above channel is referred to. And performance information comprising a primary code decoding means for decoding each of the extracted information as a plurality of track data belonging to each channel, and further for decoding the file as track data of the file. Transmission system.

Thirdly, track data belonging to a predetermined channel including at least an event consisting of pitch information, sound intensity information, and other information and information of relative time between the events, In a performance information transmitting method of receiving an input of a performance information file that the user has, compressing and transmitting the performance information file, the primary code generating means
The track data of the above performance information file is separated into a plurality of track data belonging to each channel, and the track data belonging to each channel is further separated for each of the above information while retaining the information indicating the belonging to the channel. Then, the separated pieces of information are grouped for each type of information regardless of belonging to the channel, and are arranged in independent areas for each type to generate a primary code, and the secondary code generation means is used. , The secondary code generated by the primary code generation means is compressed by a predetermined compression method with a reproduction player in the form of a program for reproducing each piece of information arranged in each area of the primary code and the performance information file. Is generated and the secondary code is transmitted by the transmission means.

Fourthly, track data belonging to a predetermined channel, which includes at least an event consisting of pitch information, sound intensity information, and other information, and information of relative time between the events, is assigned. A performance in which a performance information file that the player has and a playback player in a program form for reproducing the performance information file are compressed to decode a generated secondary code into a primary code, and further the primary code is decoded into a performance information file. In the information transmission method, by the transmission means,
The secondary code is received, the secondary code decoding means decodes the secondary code into a playback player in the form of a program for playing back the primary code and the performance information file, and the primary code decoding means performs the above. Regarding the primary code, each information for each type arranged in an independent area is extracted for each type, and referring to the information indicating the attribution to the channel,
A performance information transmission method, characterized in that each extracted information is configured as a plurality of track data belonging to each channel, and is further configured as file track data to be decoded into a file.

Fifth, the track data belonging to a predetermined channel, which includes at least an event consisting of pitch information, sound intensity information, and other information, and information of relative time between the events, is assigned. In a performance information transmission program which receives an input of a performance information file having and compresses and transmits the performance information file, the computer divides the track data of the performance information file into a plurality of track data belonging to each channel, The track data belonging to each channel is further separated for each of the above information while retaining the information indicating the attribution to the channel, and the separated information is classified for each information type regardless of the attribution to the channel. And a step of generating a primary code by disposing each type in an independent area, and the primary code generating means Generating a secondary code by compressing a reproduction player in a program form for reproducing each piece of information arranged in each area of the generated primary code and the performance information file by a predetermined compression method; And a performance information transmission program for executing the step of transmitting the next code.

[0471] Sixth, track data belonging to a predetermined channel including at least an event consisting of pitch information, sound intensity information and other information and information of relative time between the events, A performance in which a performance information file that the player has and a playback player in a program form for reproducing the performance information file are compressed to decode a generated secondary code into a primary code, and further the primary code is decoded into a performance information file. In the information transmission program, a step of receiving the secondary code by a computer, a step of decoding the secondary code by a playback player in a program form for playing back the primary code and the performance information file, and the primary step With respect to the code, each type of information placed in an independent area is extracted for each type, and information indicating the attribution to the above channel is extracted. , A performance information transmission program for executing the step of decoding each extracted information as a plurality of track data belonging to each channel, and further by configuring it as the track data of a file, Is.

[0472]

As described in detail above, according to the present invention,
Even after the decoding, the order of each performance information is restored to prevent the situation in which the decoded performance information does not match the original performance information, and the recording of some performance information is omitted. A performance information compression device, a performance information decoding device, and a performance information compression method capable of reducing the size of the next code and further efficiently compressing and expanding the data amount of the performance information in combination with a general-purpose compression method. , Performance information decoding method,
A performance information compression program and a performance information decoding program can be provided.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a performance information compression device 1 according to a first embodiment of the present invention.

FIG. 2 is a block diagram showing a detailed configuration of a primary code generation unit 2 in the performance information compression device 1 according to the first embodiment.

FIG. 3 is a block diagram showing a detailed configuration of a track separation unit 7 in the primary code generation unit 2 of the performance information compression device 1 according to the first embodiment.

FIG. 4 is a flowchart for explaining a process related to track separation by a track separation unit 7.

FIG. 5 is a block diagram showing a detailed configuration of a channel separation and channel number code generation unit 8 in the primary code generation unit 2 of the performance information compression device 1 according to the first embodiment.

FIG. 6 is a flowchart for explaining a process related to channel separation and channel number code generation by a channel separation and channel number code generation unit 8.

7A is a diagram showing a structure of track x data, FIG. 7B is a diagram showing track x data composed of a single channel in which the track x data is separated, and FIG. FIG. 8 is a diagram for explaining an effect of a process related to channel number code generation by the channel number code generation unit 8.

FIG. 8 is a block diagram showing a detailed configuration of a code generation unit 9 other than channel numbers of the performance information compression apparatus 1 according to the first embodiment.

FIG. 9 is a flowchart for explaining a coding process by the code generation unit 9 other than the channel number.

FIG. 10 is a Tr executed in step S36 of FIG.
9 is a flowchart for explaining the process related to the coding of ack [I] [J] in more detail.

11 is a flowchart for explaining in more detail the processing related to the encoding for each event, which is executed in step S47 of FIG.

FIG. 12 is a performance information compression device 1 according to the first embodiment.
3 is a block diagram showing a detailed configuration of a code arrangement unit 10 in the primary code generation unit 2 of FIG.

FIG. 13 is a flow chart for explaining a code arrangement process by the code arrangement unit 10.

FIG. 14 is a flowchart for explaining in more detail the processing relating to the Δtime code arrangement executed in step S61 of FIG.

FIG. 15 is executed in step S62 of FIG.
It is a flow chart for explaining the process concerning Note On pitch code arrangement in more detail.

16 is executed in step S63 of FIG.
It is a flow chart for explaining processing concerning Note Off pitch code arrangement in more detail.

FIG. 17 is a flowchart for explaining in more detail the processing related to velocity value code arrangement executed in step S64 of FIG.

FIG. 18 is a flowchart for explaining in more detail the processing relating to the differential bend value code arrangement executed in step S65 of FIG.

FIG. 19 is a flowchart for explaining the process relating to the differential expression value code arrangement executed in step S66 of FIG. 13 in more detail.

FIG. 20 is a flowchart for explaining in more detail the processing related to event type identification code arrangement executed in step S67 of FIG.

FIG. 21 is a diagram showing an example of an event type identification code table.

22 is a flowchart for explaining in more detail the processing relating to the channel number code arrangement executed in step S68 of FIG.

FIG. 23 is a flowchart for explaining the process relating to the other information code arrangement executed in step S69 of FIG. 13 in more detail.

24 is a diagram showing the basic arrangement order of the primary codes 5 generated by the primary code generation unit 2. FIG.

25A is a diagram showing a detailed internal arrangement order of each code of the primary code 5 generated by the primary code generation unit 2, FIG.
(B) is a figure which shows the internal arrangement | positioning order of a channel number code | symbol.

FIG. 26 is a block diagram showing a configuration of a performance information decoding device 50 according to the first embodiment of the present invention.

FIG. 27 is a performance information compression device 5 according to the first embodiment.
FIG. 3 is a block diagram showing a detailed configuration of a primary code decoding unit 51 in 0.

FIG. 28 is a performance information compression device 1 according to the first embodiment.
3 is a block diagram showing a detailed configuration of a channel-specific code extraction unit 53 in the primary code decoding unit 51 of FIG.

FIG. 29 is a flowchart for explaining a process related to code extraction by the channel-specific code extraction unit 53.

FIG. 30 is a flowchart for explaining the process related to Δ time code extraction executed in step S161 of FIG. 29 in more detail.

FIG. 31 is a flowchart for explaining in more detail the Note On pitch code extraction processing executed in step S162 of FIG. 29.

32 is a flowchart for explaining in more detail the processing related to Note Off pitch code extraction executed in step S163 of FIG. 29. FIG.

FIG. 33 is a flowchart for explaining the processing relating to velocity value code extraction executed in step S164 of FIG. 29 in more detail.

FIG. 34 is a flowchart for explaining in more detail the processing relating to differential bend value code extraction executed in step S165 of FIG.

FIG. 35 is a flow chart for explaining the processing relating to the differential expression value code extraction executed in step S166 of FIG. 29 in further detail.

FIG. 36 is a flow chart for explaining the process relating to the event type identification code extraction executed in step S167 of FIG. 29 in more detail.

FIG. 37 is a flowchart for explaining in more detail the processing related to channel number code extraction executed in step S168 of FIG.

FIG. 38 is a flow chart for explaining the process relating to the extraction of other information code executed in step S169 of FIG. 29 in more detail.

FIG. 39 is a performance information compression device 5 according to the first embodiment.
5 is a block diagram showing a detailed configuration of a track decoding unit 54 in the 0 primary code generation unit 52. FIG.

FIG. 40 is a flowchart for explaining a decoding process by the track decoding unit 54.

41 is a flowchart for explaining in more detail the processing relating to Track [I] [J] decoding executed in step S266 of FIG. 41. FIG.

42 is a flowchart for explaining in more detail the process related to the decoding for each event type, which is executed in step S273 in FIG. 41. FIG.

FIG. 43 is a performance information compression device 5 according to the first embodiment.
5 is a block diagram showing a detailed configuration of a track merging and channel number code decoding unit 55 in the 0 primary code generation unit 52. FIG.

[Fig. 44] Fig. 44 is a flowchart for describing a process related to track merge and channel number code decoding by the track merge and channel number code decoding unit 55.

FIG. 45 is a flowchart for explaining in more detail the processing related to track merging executed in step S305 of FIG. 44.

FIG. 46 is a performance information compression device 5 according to the first embodiment.
5 is a block diagram showing a detailed configuration of a track arrangement unit 56 in the 0 primary code generation unit 52. FIG.

FIG. 47 is a flowchart for explaining a process related to track arrangement by the track arrangement unit 56.

FIG. 48 is a performance information compression device 7 according to the second embodiment.
3 is a block diagram showing a detailed configuration of a code generation unit 100 other than a channel number of 0. FIG.

49 is a flowchart for explaining in more detail the processing related to the encoding for each event, which is executed in step S47 of FIG. 10 in the second embodiment.

FIG. 50 is a flowchart for explaining a code arrangement process by the code arrangement unit 10.

51 is a flow chart for explaining the processing relating to the Δtime code arrangement executed in step S381 of FIG. 50 in more detail.

52 is a flow chart for explaining the process relating to pitch code arrangement executed in step S382 of FIG. 50 in more detail.

53 is a flow chart for explaining the process relating to gate time pitch code arrangement executed in step S383 of FIG. 50 in further detail.

54 is a flow chart for explaining the processing relating to the velocity value code arrangement executed in step S384 of FIG. 50 in further detail.

FIG. 55 is a flowchart for explaining the process relating to the differential bend value code arrangement executed in step S385 of FIG. 50 in more detail.

FIG. 56 is a flowchart for explaining the processing relating to the differential volume value code arrangement executed in step S386 of FIG. 50 in more detail.

57 is a flow chart for explaining the process relating to the event type identification code arrangement executed in step S387 of FIG. 50 in more detail.

FIG. 58 is a diagram showing an example of an event type identification code table.

FIG. 59 is a flowchart for explaining in more detail the processing relating to the channel number code arrangement executed in step S388 of FIG. 50.

FIG. 60 is a flow chart for explaining the process relating to the other information code arrangement executed in step S389 of FIG. 50 in more detail.

FIG. 61 is a diagram showing a basic arrangement order of the primary codes 5 generated by the primary code generation unit 2 in the second embodiment.

FIG. 62 is a performance information decoding device 8 according to the second embodiment.
3 is a block diagram showing a detailed configuration of a channel-specific code extraction unit 110 adopted by 0. FIG.

[Fig. 63] Fig. 63 is a flowchart for explaining a process related to code extraction by the channel-specific code extraction unit 110.

FIG. 64 is a flowchart for explaining in more detail the process related to Δ time code extraction executed in step S481 of FIG. 63.

65 is a flow chart for explaining the process relating to pitch code extraction executed in step S482 of FIG. 63 in further detail.

66 is a flow chart for explaining the process relating to gate time code extraction executed in step S483 of FIG. 63 in further detail.

67 is a flow chart for explaining the process relating to velocity value code extraction executed in step S484 of FIG. 63 in further detail.

FIG. 68 is a flowchart for explaining in more detail the processing relating to differential bend value code extraction executed in step S485 of FIG. 63.

69 is a flow chart for explaining the process relating to the differential volume value code extraction executed in step S486 of FIG. 63 in further detail.

70 is a flowchart for explaining in more detail the processing relating to the event type identification code extraction executed in step S487 of FIG. 63. FIG.

71 is a flowchart for illustrating in more detail the processing relating to channel number code extraction executed in step S488 of FIG. 63. FIG.

72 is a flow chart for explaining the process relating to the extraction of other information code executed in step S489 of FIG. 63 in further detail.

FIG. 73 is a performance information decoding device 8 according to the second embodiment.
3 is a block diagram showing a detailed configuration of a track decoding unit 120 in 0. FIG.

[Fig. 74] Fig. 74 is a flowchart for explaining in more detail the process related to the decoding for each event type, which is executed in step S273 of Fig. 41 in the second embodiment.

FIG. 75 is a diagram showing a first configuration example of a system according to a third embodiment.

FIG. 76 is a diagram showing a second configuration example of the system according to the third embodiment.

FIG. 77 is a diagram showing a third configuration example of the system according to the third embodiment.

FIG. 78 is a diagram showing a fourth configuration example of the system according to the third embodiment.

79 is a diagram showing a type correspondence table for each performance information format. FIG.

[Explanation of symbols]

1 Performance information compression device 2 Primary code generator 3 Secondary code generator 4 music files 5 Primary code 6 secondary code 7 Track separation section 7a file analysis section 8 channel separation and channel number code generator 8a track analysis section 8b channel number code generator 9 Code generator other than channel number 10 Code Arrangement Unit 11 track data 21 Event type analysis and identification section 22 Δ time code generator 23 Note On Pitch code generator 24 Note Off Pitch code generator 25 Velocity value code generator 26 Differential Bend Value Code Generator 27 Differential Expression Value Code Generator 28 Event type identification code generator 29 Other information code generator 50 Performance information decoding device 51 Secondary code decoding unit 52 primary code decoding unit 53 Channel-specific code extraction unit 54 track decoder 55 Track merge and channel number code decoder 55a track merge section 55b channel number code decoding unit 56 Track placement section 60 Δ time code decoding unit 61 Note On Pitch code decoding unit 62 Note Off pitch code decoding unit 63 Velocity value code decoding unit 64 Differential Bend Value Code Decoding Unit 65 Differential Expression Value Code Decoding Unit 66 Event type identification code decoding unit 67 Other information code decoding unit 68 Event writing department 69 Track data consisting of a single channel 70 Performance information compression device 80 Performance information decoding device Code generator other than 100 channel numbers 101 Event type analysis and identification section 102 Δ time code generator 103 pitch code generator 104 Gate time code generator 105 Velocity value code generator 106 differential bend value code generator 107 differential volume value code generation unit 108 Event type identification code generator 109 Other information code generator 110 Code Extractor for Each Channel 120 Code extractor other than channel number 121 Δ time code extraction unit 122 pitch code extraction unit 123 Gate time code extraction unit 124 Velocity value code extraction unit 125 Differential Bend Value Code Extraction Unit 126 Differential Volume Value Code Extraction Unit 127 Event type identification code extraction unit 128 Other information code extraction unit 129 Event writing department

─────────────────────────────────────────────────── ─── Continuation of the front page (56) Reference JP-A-9-16168 (JP, A) JP-A-9-153819 (JP, A) JP-A-9-22285 (JP, A) JP-A-2002-91436 (JP, A) JP 2002-91437 (JP, A) JP 8-152881 (JP, A) JP 8-202358 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G10H 1/00-7/12 H03M 7/46

Claims (30)

(57) [Claims] 1. An event consisting of at least information on a sounding start pitch, information on a sound intensity, information on a sounding end pitch, other information, and information on a relative time between the events.
Receives the file having a plurality of track data that are, in the performance information compression apparatus for compressing the file, a plurality of track data of the file is separated into belt each rack data attributable to each channel, decoding Sometimes come
While holding the information showing attribution to the channel to illuminate ,
The information track data belonging to each channel further separated for each type, summary information of the separated each type for each different species, to produce a primary code by placing in a separate area for each different species A primary code generating means, and a secondary code generating means for compressing each information arranged in each area of the primary code generated by the primary code generating means by a predetermined compression method. Performance information compression device. 2. A plurality of events including at least pitch information, sound intensity information, sound length information, and other information, and relative time information between the events .
Receives the file with the track data, the performance information compression apparatus for compressing the file, a plurality of track data of the file is separated into belt rack every data attributable to each channel, when the decoding three
While holding the information showing attribution to the channel to illuminate ,
The information track data belonging to each channel further separated for each type, summary information of the separated each type for each different species, to produce a primary code by placing in a separate area for each different species A primary code generating means, and a secondary code generating means for compressing each information arranged in each area of the primary code generated by the primary code generating means by a predetermined compression method. Performance information compression device. 3. The plurality of events when the primary code generation means determines that a plurality of events related to the same time are recorded in the track data based on information of relative time between the events. For the event recorded at the end of the event group consisting of, or when it is determined that there is a single event related to a certain time, exclude that event from the target that retains information indicating attribution to the channel. The performance information compression apparatus according to claim 1 or 2, further comprising: Wherein the primary code generating means assigns an event type identification code for each event included in the track data, collectively the event type identification code, comprising further placing a separate area Claim 1 characterized by
Alternatively, the performance information compression device described in 2. 5. The primary code generation means uses at least the event type identification information of the most frequently used event among the events included in the event group recorded in the performance information file as a relative of the primary code. 3. The performance information compression apparatus according to claim 1, further comprising recording the time information by including it in the time code. 6. The primary code generation means is for the event.
Is a given type, it indicates attribution to the channel.
To record event class information instead of
The performance information compression apparatus according to claim 2, which is further characterized. 7. The performance information compression apparatus according to claim 1.
Decompressed and generated secondary code by a predetermined decoding method
Decode to a primary code, and then decode the primary code to a file
In the performance information decoding device, a secondary code decoding means for decoding the secondary code into a primary code
And the above primary codes are arranged in independent areas.
Extract each information for each type into the above channel
While referring to the information that indicates the attribution of
Configured as multiple track data belonging to the channel,
Furthermore, by configuring it as the track data of the file,
And a primary code decoding means for decoding into a file.
Characteristic performance information decoding device. 8. The performance information compression apparatus according to claim 2.
Decompressed and generated secondary code by a predetermined decoding method
Decode to a primary code, and then decode the primary code to a file
In the performance information decoding device, a secondary code decoding means for decoding the secondary code into a primary code
And the above primary codes are arranged in independent areas.
Extract each information for each type into the above channel
While referring to the information that indicates the attribution of
Configured as multiple track data belonging to the channel,
Furthermore , by configuring it as the track data of the file,
And a primary code decoding means for decoding into a file.
Characteristic performance information decoding device. 9. The primary code decoding means is adapted to perform the above decoding.
The event types are arranged in separate areas.
Event type identification assigned to each event
A further feature of identifying based on a code
Item 7. The performance information decoding device according to item 7 or 8. 10. The event of the primary code decoding means is
If it is of a certain type, please attribute to the above channel.
Refer to the event class information instead of the information shown
Claims, characterized by the fact that they constitute rack data
Item 8. The performance information decoding device according to item 8. 11. Information on at least sounding start pitch and sound
It consists of strength information, end tone pitch information, and other information.
Event and the information of the relative time between the events
Input of a file with multiple track data
Receive the performance information compression method that compresses the file
Te, by the primary code generating unit, a plurality of tracks of the file
For each track data belonging to each channel
Separation and attribution to the channel for reference when decoding
While holding the information indicating
The data of the data is further separated by type, and this separated
Information that is classified by type is grouped by type, and an independent area for each type
To generate a primary code by arranging the primary code and the secondary code generating means, and the primary code generating means.
Predetermine each information placed in each area of the generated primary code
And a step of compressing the performance information by the above compression method. 12. At least pitch information and sound intensity information
Event consisting of information, sound length information, and other information
And a plurality of information including relative time information between the events.
Input the file containing the track data of
In a performance information compression method for compressing a file, a plurality of tracks of the file are
For each track data belonging to each channel
Separation and attribution to the channel for reference when decoding
While holding the information indicating
The data of the data is further separated by type, and this separated
Information that is classified by type is grouped by type, and an independent area for each type
To generate a primary code by arranging the first code and the secondary code generation means, and the primary code generation means
Predetermine each information placed in each area of the generated primary code
And a step of compressing the performance information by the above compression method. 13. The primary code generation means uses the track based on information of relative time between the events.
Data, multiple events related to the same time are recorded.
If it is determined that the event consisting of the above multiple events
The event recorded at the end of the group
If it is determined that there is a single event related to a fixed time,
About the event, information showing attribution to the channel
The method according to claim 11 or 12, further comprising a step of excluding the object from being held.
Method of compressing performance information in the series. 14. An event type identification code is included in the track data by the primary code generation means.
Allocate to each event and assign the event type identification code
It also has a step of collectively arranging them in independent areas.
The performance according to claim 11 or 12, characterized in that
Information compression method. 15. An event group recorded in a performance information file by the primary code generating means.
Of the most frequently used events
The event type identification information as the relative time of the primary code
13. The method according to claim 11 or 12, further comprising the step of recording by including the code.
Method of compressing performance information in the series. 16. A channel if the event is of a predetermined type by the primary code generation means.
Event class information is recorded instead of information indicating attribution to
The performance according to claim 12, further comprising a step of recording.
Information compression method. 17. A performance information compression method according to claim 11.
The secondary code compressed and generated by
To the primary code and then to the file.
In a performance information decoding method for decoding, the secondary code decoding means restores the secondary code to a primary code.
And a step of decoding the primary code independently by the primary code decoding means.
Information for each type that is placed in the specified area for each type
And refer to the information showing attribution to the above channels.
The multiple extracted information is assigned to each channel.
Track data of the file.
To decrypt into a file by configuring it as a
And a performance information decoding method comprising: 18. A performance information compression method according to claim 12.
The secondary code compressed and generated by
To the primary code and then to the file.
In a performance information decoding method for decoding, the secondary code decoding means restores the secondary code to a primary code.
And a step of decoding the primary code independently by the primary code decoding means.
Information for each type that is placed in the specified area for each type
And refer to the information showing attribution to the above channels.
The multiple extracted information is assigned to each channel.
Track data of the file.
To decrypt into a file by configuring it as a
And a performance information decoding method comprising: 19. The decoding means for decoding by the primary code decoding means.
Issue, each event located in a separate area.
Event type assigned to each event
The method further includes the step of identifying based on the type identification code.
Performance information according to claim 17 or 18, characterized in that
Decryption method. 20. An event is provided by the primary code decoding means.
Channel is of a given type, the
Refer to event class information instead of the information indicating the genus
Further comprising the step of constructing one track data
19. The performance information decoding method according to claim 18, wherein: 21. At least information on a note starting pitch and a note
It consists of strength information, end tone pitch information, and other information.
Event and the information of the relative time between the events
Input of a file with multiple track data
Performance information compression program that receives and compresses the file
In the computer, multiple track data of the above file can be
Separate each track data belonging to it and refer to it when decrypting.
While holding the information showing attribution to the channel to illuminate,
Track data information belonging to each channel for each type
It is further separated, and the separated information for each type is divided into each type.
First, by arranging it in an independent area for each type, the primary code
And the information arranged in each area of the generated primary code.
A performance information compression program for executing a step of compressing by a predetermined compression method . 22. At least pitch information and sound intensity information
Event consisting of information, sound length information, and other information
And a plurality of information including relative time information between the events.
Input the file containing the track data of
In a performance information compression program that compresses a file, a computer is used to store multiple track data of the above file for each channel.
Separate each track data belonging to it and refer to it when decrypting.
While holding the information showing attribution to the channel to illuminate,
Track data information belonging to each channel for each type
It is further separated, and the separated information for each type is divided into each type.
First, by arranging it in an independent area for each type, the primary code
And the information arranged in each area of the generated primary code.
A performance information compression program for executing a step of compressing by a predetermined compression method . 23. Based on information on relative time between the events , the computer is allowed to track the track.
Data, multiple events related to the same time are recorded.
If it is determined that the event consisting of the above multiple events
The event recorded at the end of the group
If it is determined that there is a single event related to a fixed time,
About the event, information showing attribution to the channel
23. The performance information according to claim 21 or 22, further comprising the step of excluding the object from the holding target.
Compression program. 24. The computer includes an event type identification code in the track data.
Allocate to each event and assign the event type identification code
Collectively perform steps to place in independent areas
23. Performance information according to claim 21 or 22, characterized in that
Compression program. 25. The computer includes a group of events recorded in a performance information file.
Of the most frequently used events
The event type identification information as the relative time of the primary code
23. The performance information according to claim 21 or 22, further comprising the step of recording by including the code.
Compression program. 26. A channel to the computer if the event is of a predetermined type
Event class information is recorded instead of information indicating attribution to
A contract characterized by further executing the step of recording
The performance information compression program according to claim 22. 27. A performance information compression program according to claim 21.
Predetermined decoding of secondary code generated by compression with gram
Method to the primary code and then the primary code
In the performance information decoding program for decoding the primary code, the step of decoding the secondary code into the primary code and the primary code are arranged in independent areas in the computer .
Extract each information for each type into the above channel
While referring to the information that indicates the attribution of
Configured as multiple track data belonging to the channel,
Furthermore, by configuring it as the track data of the file,
And a performance information decoding program characterized by executing the steps of
Mu. 28. A performance information compression program according to claim 22.
Predetermined decoding of secondary code generated by compression with gram
Method to the primary code and then the primary code
In the performance information decoding program for decoding the primary code, the step of decoding the secondary code into the primary code and the primary code are arranged in independent areas in the computer .
Extract each information for each type into the above channel
While referring to the information that indicates the attribution of
Configured as multiple track data belonging to the channel,
Furthermore, by configuring it as the track data of the file,
And a performance information decoding program characterized by executing the steps of
Mu. 29. Each of the computers arranged in independent areas at the time of the decoding.
The event type is assigned to each event.
Further identifying step based on the vent type identification code
29. The method according to claim 27 or 28, characterized in that
The performance information decoding program in the above. 30. The channel to the computer if the event is of a predetermined type.
Refer to event class information instead of information indicating attribution to
Further steps to compose track data while illuminating
29. The performance information restoration according to claim 28, wherein
No. program.