JPH08146991A - Information processing apparatus and control method thereof - Google Patents

Abstract

(57) [Abstract] [Purpose] It is possible to make a dialogue by voice input a natural state and quickly obtain necessary information for the user. [Structure] A voice input from a microphone 1 is converted into a digital signal via an A / D conversion unit 2, and then recognized by a voice recognition unit 3. A response sentence is created based on the recognized result, and the voice synthesis unit 6 converts it into a voice signal and displays it on the display unit 5. When the user utters during this output, the output is temporarily interrupted, and it is determined whether to resume or end according to the utterance content.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information processing apparatus and its control method, and more particularly to an information processing apparatus which responds to utterances and its control method.

[0002]

2. Description of the Related Art Recently, a voice dialog device has been used. A conventional voice dialogue system will be described with reference to FIG. First, the user looks into the microphone, etc.
Suppose you uttered as in (1). The voice input to the microphone is converted from an analog signal to a digital signal by the A / D conversion unit 2, and the digital signal is converted into a Japanese sentence as shown in FIG.
This Japanese sentence is interpreted by the dialogue management unit 4, and a corresponding response sentence as shown in FIG. 3B is created. This response sentence is divided into words by the voice synthesis unit 6, added with reading and accent, converted into digital signals of phonological parameters and prosody parameters, converted into analog signals by the D / A conversion unit 7, and the speaker 8 Is output as voice. By repeating the voice input and the voice output as described above, the dialogue between the user and the system is performed.

[0003]

In the above-mentioned conventional voice interaction system, when the user inputs voice during voice output,
There was no naturalness of the dialogue because the user input was ignored and the composition continued, or the user input stopped the composition.

[0004]

[Means for Solving the Problems] and

The present invention has been made in view of the above problems, and makes the dialogue by voice input natural.
Further, it is an object of the present invention to provide an information processing apparatus and a control method thereof that enable a user to quickly obtain necessary information.

To solve this problem, for example, the information processing apparatus of the present invention has the following configuration. That is, in an information processing device including a voice input unit, a voice recognition unit, and an output unit that responds to the voice input unit, a voice input from the voice input unit is recognized, and a sentence corresponding to the voice input is recognized. And an output control unit configured to form an output sentence corresponding to the sentence generated by the sentence generation unit and controlling the output unit to output the output sentence to the outside, and outputting by the output control unit, A first control unit that monitors the input from the voice input unit and temporarily stops the output when a voice input is made, and a response sentence that is being output according to the input voice. The second to control whether to restart or suspend
And the control means of.

According to an embodiment of the present invention, it is desirable that the output means includes both a voice output means and a display means for displaying a dialogue sentence. As a result, since both the input sentence and the response sentence are displayed, it is possible to surely confirm the content output by voice.

Further, the second control means may be controlled by the length of the voice information inputted from the voice input means.
As a result, it is possible to simplify the control because only the length of the voice is required.

Further, the second control means may be controlled by the recognition result of the voice information inputted from the voice input means. As a result, it becomes possible to detect the user's intention and reliably perform the corresponding processing.

Further, when the second control means decides to restart the output of the response sentence, the delimiter position is searched along the direction going back from the interrupted position, and the output is restarted from the searched position. Is desirable. In particular, the delimiter position includes one phrase, one accent, one breath paragraph, or the beginning of a sentence, so that the output sentence is output correctly and in a natural state even if the output is in the middle. It will be possible.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An example of an actual sister according to the present invention will be described in detail below with reference to the accompanying drawings.

FIG. 2 is a block diagram showing the configuration of the voice interactive apparatus of this embodiment. This device is a microphone 1, A / D
It is composed of a conversion unit, a voice recognition unit 3, a dialogue management unit 4, a display unit 5 for displaying conversation contents, a voice synthesis unit 6, a D / A conversion unit 7, and a speaker 8. Further, the CPU 9
Each of the above components is controlled according to a program stored in the main memory therein.

The processing procedure of the CPU 9 is as shown in FIG. 1. A voice synthesis output start step S11, a step S12 for determining whether or not a voice input exists, a step S13 for continuing the voice synthesis output, and a voice Step S14 for determining whether the input is completed, Step S15 for temporarily suspending the voice synthesis output, and Step S for determining whether the voice input duration is larger or smaller than a certain threshold value.
16 and step S17 for ending the voice synthesis output.

Next, the operation of the voice dialogue system of the present embodiment thus constructed will be described with reference to FIGS. 1 and 4.

First, it is assumed that the user occurs as in the procedure (1) in FIG. Then, in step S11, as shown in FIG.
The voice synthesis output is started as in the procedure (2). The voice recognition unit 3 accepts the input even during the voice synthesis output, and the result is sent to the dialogue management unit 4 at any time. The dialogue management unit 4 determines in step S12 whether the voice input is present, and the input is If it does not exist, the voice synthesis output is continued in step S13, and the process returns to step S12. If it is determined in step 14 that the voice input is completed, the process proceeds to step 16. When it is determined in step 16 that the voice input duration is smaller than the certain threshold Th, the voice synthesis output is continued in step 13 and the process returns to step 12. If it is determined in step 16 that the voice input duration is greater than a certain threshold Th,
Moving to step 17, the voice synthesis output is ended.

Now, as shown in step (2) of FIG. 4, when the system performs voice synthesis output up to “Explanation of Kobe City Suma Seaside Aquarium is 24000 square meters of floor space”, as shown in step (3) of FIG. It is assumed that the user has generated "Eh". Then, it is determined in step S12 that there is a voice input, and voice synthesis output is temporarily suspended in steps S14 and S15.

Here, in step 16, "huh"
It is assumed that it is determined that the lifetime of is less than Th. Then, in step S17, the voice synthesis output is continued as in the procedure (4) of FIG. 4, "Aquarium main building, sea otter building ...". Furthermore, it is assumed that the user has generated "Thank you" as in step (5) of FIG. 4 when the system synthesizes and outputs "horde of swimming ... . Then, it is determined in step 12 that the voice input is present, and the voice synthesis output is temporarily suspended in steps S14 and S15.
Here, it is assumed that it is determined in step 16 that the duration of "thank you" is longer than Th. Then, the process proceeds to step 17 to end the voice synthesis output. The above conversation status is displayed on the screen of the display unit 5, as shown in FIG.

As described above, according to the present embodiment, when the user's utterance is detected during the voice synthesis output based on the response sentence, the voice synthesis output is temporarily stopped and the utterance content (the above embodiment). Then, it is determined whether to continue or interrupt the speech synthesis output according to the duration. Therefore, as described above, when the user determines that there is no need for further information, it is possible to stop subsequent speech synthesis by uttering "Thank you" or the like. It is also possible to make the next inquiry immediately.

[Description of Second Embodiment] Next, a second embodiment of the present invention will be described.

Referring to FIG. 5, the process in the voice interactive system of this embodiment is performed by a voice synthesis output starting step S51,
Step S5 of determining whether or not voice input is present
2. Step S53 of continuing the voice synthesis output, step S54 of determining whether the voice input is finished, step S55 of temporarily interrupting the voice synthesis output, and determining whether the voice input duration is larger or smaller than a certain threshold. S56, step S of determining the continuation position of the voice synthesis output
57, and step S58 for terminating the voice synthesis output.
Consists of The above steps of the second embodiment are stored in advance in the main memory of the CPU 9.

Next, the operation of the voice interactive system of the present embodiment thus constructed will be described with reference to FIGS. 5 and 6.

First, it is assumed that the user utters as in step (1) of FIG. On the other hand, the system is like the procedure (2) in Fig. 6, "The explanation of Kobe City Suma Sea Aquarium is 2
It is assumed that at the stage of voice synthesis output up to 4000 square meters, the user utters “um” as in step (3) of FIG. 6, and it is determined that the existence period is smaller than Th. Then, in step S57, the restart position of the speech synthesis output is set to the beginning position of the phrase one phrase before the interrupted phrase (however, the restart position is a position with a good break, such as one accent phrase before or one breath paragraph before). If so any).

Next, in step 53, the system continues the speech synthesis output as "Procedure of the aquarium on the site of 24,000 square meters ..." As in step (4) of FIG. In addition, as shown in step (4) in Fig. 6, the system says "The horde of ... swims."
It is assumed that the user has generated "Thank you" as shown in FIG. 6 (5) at the time of performing voice synthesis output up to. Then, it is determined in step 52 that the voice input is present, and the voice synthesis output is temporarily interrupted in steps S54 and S55. Here, it is assumed that it is determined in step S56 that the duration of "thank you" is longer than Th. Then, the process proceeds to step S58 to end the voice synthesis output. The above conversation status is displayed on the screen of the display unit 5 as shown in FIG.

As a result of the above, according to the second embodiment, when the voice synthesis output is temporarily interrupted, the phrase immediately before the interruption or the accent or the like is heard not from the interruption position. Since the operation is restarted from the easy position, it becomes user-friendly as compared with the first embodiment.

[Description of Third Embodiment] Next, a third embodiment of the present invention will be described.

Referring to FIG. 7, the process in the voice interactive system of this embodiment is performed by a voice synthesis output starting step S71.
Step S7 of determining whether or not voice input is present
2. Step S73 of continuing voice synthesis output, step S74 of determining whether voice input has ended, step S75 of temporarily suspending voice synthesis output, step S76 of determining whether the voice recognition result is positive or negative, Step S77 of deciding the continuous position of the voice synthesis output and step S78 of terminating the voice synthesis output.

Next, the operation of the voice dialogue system of the present embodiment thus constructed will be described with reference to FIGS. 7 and 8.

First, it is assumed that the user occurs as in the procedure (1) of FIG. It is assumed that the system responds to this in response to "Seven buildings such as Kobe City ... scattered" as in step (2) of FIG. At this point, if the user answers "Yes" as in step (3) of FIG. 8, it is determined in step 72 that the voice input is present, the voice synthesis output is temporarily suspended in step 75, and the voice input "in step 76". Whether the recognition result of "Yes" is positive or negative is determined. Here, "Yes", "No", "Understood", etc. are regarded as affirmative, and the process proceeds to step 77, where the continuous position of the voice synthesis output is determined. Here, since it is at the end of the sentence, it is considered that the division is good, the process proceeds to step 73, and the voice synthesis output is continued, and as in step (4) of FIG.
A horde of ... swims and is a sea otter. ”The voice synthesis output continues. At this point, if the user makes a non-affirmative occurrence of "please end" as in step (5) of FIG. 8, it is determined in step 76 that the recognition result is negative, and the flow proceeds to step 78 to synthesize voice output. To finish. The above conversation status is displayed on the screen of the display unit 5, as shown in FIG.

As described above, according to this embodiment, a voice recognition means for receiving an input at any time, a voice synthesizing means for outputting a response to the input by voice, and a dialogue managing means for managing input / output are generated. The naturalness of the dialogue between the user and the system is improved by providing the means for controlling the response by the voice synthesis of the response sentence according to the user's input.

The present invention may be applied to a system including a plurality of devices or an apparatus including a single device. Further, it goes without saying that the present invention can be applied to the case where it is achieved by supplying a program to a system or an apparatus.

[0030]

As described above, according to the present invention, it is possible to make a dialogue by voice input a natural state and to obtain necessary information quickly for the user.

[0031]

[Brief description of drawings]

FIG. 1 is a flowchart showing a process of a voice interaction system according to a first embodiment of this invention.

FIG. 2 is a block diagram showing an example of the configuration of a voice dialog device to which the voice dialog system of the present embodiment is applied.

FIG. 3 is a diagram showing an operation example of a conventional voice interaction method.

FIG. 4 is a diagram showing an operation example of the first exemplary embodiment of the present invention.

FIG. 5 is a flow chart showing processing of the second exemplary embodiment of the present invention.

FIG. 6 is a diagram showing an operation example of the second exemplary embodiment of the present invention.

FIG. 7 is a flow chart showing the processing of the third exemplary embodiment of the present invention.

FIG. 8 is a diagram showing an operation example of the third exemplary embodiment of the present invention.

[Explanation of symbols]

 1 Microphone 2 A / D Converter 3 Voice Recognition 4 Dialog Management 5 Display 6 Voice Synthesizer 7 D / A Converter 8 Speaker

Claims (12)

[Claims] 1. An information processing apparatus comprising a voice input means, a voice recognition means, and an output means for responding to the voice input means, wherein the voice input from the voice input means is recognized and used for voice input. A sentence generation unit that generates a corresponding sentence; an output control unit that configures an output sentence corresponding to the sentence generated by the sentence generation unit and controls the output unit to output the sentence to the outside; During the output, the input from the voice input means is monitored, and when there is a voice input, the first control means that temporarily stops the output and the output is in progress according to the input voice. An information processing apparatus comprising: a second control unit that controls whether to restart or suspend the response sentence. 2. The information processing apparatus according to claim 1, wherein the output unit includes both a voice output unit and a display unit for displaying a dialogue sentence. 3. The information processing apparatus according to claim 1, wherein the second control unit controls according to a length of voice information input from the voice input unit. 4. The information processing apparatus according to claim 1, wherein the second control unit controls according to a recognition result of voice information input from the voice input unit. 5. The second control means, when determining to restart the output of the response sentence, searches for a delimiter position along a direction going back from the interrupted position, and restarts the output from the searched position. The information processing apparatus according to claim 1, wherein the information processing apparatus comprises: 6. The information processing apparatus according to claim 1, wherein the delimiter position includes one phrase, one accent, one breath paragraph, or the beginning of a sentence. 7. A method of controlling an information processing apparatus, comprising: a voice input unit, a voice recognition unit, and an output unit that responds to the voice input unit, wherein a voice input from the voice input unit is recognized, A sentence generation step of generating a sentence corresponding to a voice input; an output control step of forming an output sentence corresponding to the sentence generated in the sentence generation step and controlling the output means to output to the outside; During the output in the control step, the input from the voice input means is monitored, and when there is a voice input, the first control step of temporarily stopping the output, and the middle of the output depending on the input voice And a second control step for controlling whether to restart or suspend the response sentence that was described above. 8. The method according to claim 7, wherein the output means includes both a voice output means and a display means for displaying a dialogue sentence. 9. The method of controlling an information processing apparatus according to claim 7, wherein the second control step is controlled by the length of the voice information input from the voice input means. 10. The control method of an information processing apparatus according to claim 7, wherein the second control step is controlled by a recognition result of voice information input from a voice input unit. 11. When the second control step determines to restart the output of the response sentence, the delimiter position is searched along a direction going back from the interrupted position, and the output is restarted from the searched position. The method for controlling an information processing device according to claim 7, wherein 12. The control method of the information processing apparatus according to claim 7, wherein the delimiter position includes one phrase, one accent, one breath paragraph, or the beginning of a sentence.