KR102268245B1 - Device, method and server for providing voice recognition service - Google Patents

Abstract

A terminal providing a voice recognition service according to an embodiment of the present invention includes a sound input unit receiving a sound signal through each of a first microphone disposed on one side of the terminal and a second microphone disposed on the other side of the terminal, the first microphone A sound pressure level measuring unit that measures sound pressure levels for each of the first sound signal input through the first sound signal and the second sound signal input through the second microphone, and one of the first microphone and the second microphone based on the measured sound pressure level and a microphone selector for receiving a voice command that selects as a microphone for receiving a voice command for receiving a voice command signal, and a voice recognition performing unit that performs voice recognition based on a voice command signal input through the microphone for receiving the selected command.

Description

Terminal, method and server for providing voice recognition service {DEVICE, METHOD AND SERVER FOR PROVIDING VOICE RECOGNITION SERVICE}

The present invention relates to a terminal, a method and a server for providing a voice recognition service.

The voice recognition apparatus may start a voice recognition service based on a voice wake-up method. For example, when a voice command signal including a wake up word is input, the voice recognition device prepares for voice recognition according to the call word, and provides a voice recognition service according to the voice command signal input through a microphone can do.

On the other hand, when a specific noise is also input when the call word (or voice command) is input to the microphone (microphone) of the voice recognition device, this is a factor that makes it difficult to recognize the call word (or voice command). do. For example, if a specific sound source is being output through the speaker of the voice recognition device or when the voice recognition device receives a voice command signal including a calling word in a noisy environment around the voice recognition device, it may not be able to recognize the calling word this is high

In order to solve this problem, an echo that is a phenomenon in which the self-reproduced sound of the voice recognition device is re-inputted into the microphone using an echo canceller (eg, Acoustic Echo Canceller (AEC)) or by using a noise suppression (NS) function Techniques for reducing noise are being used.

However, the voice recognition device has both a microphone and a speaker due to its characteristics, and in particular, in a specific device such as a smart phone, the microphone and the speaker may be disposed adjacently. In this case, since the sound source output from the speaker acts as a very large noise, it is difficult to solve the above-described problem using the techniques presented above.

For example, when a microphone and a speaker of the voice recognition device are disposed adjacent to each other, even if a user utters a call word in a situation where a specific sound source is output through the speaker, the sound source and the call word are input together into the microphone, so voice recognition The device often fails to respond to the caller.

Japanese Patent Publication No. 5862318 (registered on Jan. 8, 2016)

In general, the voice recognition apparatus is provided with a plurality of microphones in consideration of the directionality of the spoken voice. The present invention is to solve the above-mentioned problem by selecting a speaker having a low sound pressure level as a microphone for receiving a voice command to perform voice recognition in such a voice recognition apparatus when a user's voice command signal is received.

The present invention However, the technical problems to be achieved by this embodiment are not limited to the technical problems as described above, and other technical problems may exist.

As a technical means for achieving the above technical problem, a terminal providing a voice recognition service receives a sound signal through each of a first microphone disposed on one side of the terminal and a second microphone disposed on the other side of the terminal. input unit; a sound pressure level measuring unit for measuring sound pressure levels for each of the first sound signal input through the first microphone and the second sound signal input through the second microphone; a microphone selection unit for receiving a voice command for selecting one of the first microphone and the second microphone as a microphone for receiving a voice command for receiving a voice command signal based on the measured sound pressure level; and a voice recognition performing unit configured to perform voice recognition based on the voice command signal input through the selected command receiving microphone.

A method of providing a voice recognition service according to another embodiment of the present invention includes: receiving a sound signal through each of a first microphone disposed on one side of a terminal and a second microphone disposed on the other side of the terminal; measuring a sound pressure level for each of a first sound signal input through the first microphone and a second sound signal input through the second microphone; selecting one of the first microphone and the second microphone as a microphone for receiving a voice command for receiving a voice command signal based on the measured sound pressure level; and performing voice recognition based on the voice command signal input through the selected command receiving microphone.

A server providing a voice recognition service according to another embodiment of the present invention includes a sound receiving unit that receives a sound signal input through each of a first microphone and a second microphone of the terminal from a terminal, the first microphone A sound pressure level measuring unit for measuring sound pressure levels for each of the first sound signal input through the first sound signal and the second sound signal input through the second microphone, the first microphone and the second microphone based on the measured sound pressure level A voice command receiving microphone selector selecting one of the microphones as a voice command receiving microphone for receiving a voice command signal, and voice recognition performing voice recognition based on the voice command signal input through the selected command receiving microphone includes an execution unit.

According to any one of the above-described problem solving means of the present invention, the present invention is a microphone for receiving a voice command for receiving a voice command signal from among the plurality of microphones based on the sound pressure level for each sound signal input through each of the plurality of microphones , and performing voice recognition based on a voice command signal input through the selected microphone, the voice recognition apparatus may easily respond to the call word.

Further, according to the present invention, it is possible to provide a stable voice recognition service even in a noisy environment by improving the voice recognition performance of the voice recognition apparatus.

1 is a block diagram of a system for providing a voice recognition service according to an embodiment of the present invention.
2 is a block diagram of the voice recognition service providing terminal shown in FIG. 1 according to an embodiment of the present invention.
3 is a view for explaining a method of selecting a microphone for receiving a voice command for receiving a voice command signal according to an embodiment of the present invention.
4 is a flowchart illustrating a method of providing a voice recognition service according to an embodiment of the present invention.
5 is a block diagram of the voice recognition server shown in FIG. 1 according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily implement them. However, the present invention may be embodied in many different forms and is not limited to the embodiments described herein. And in order to clearly explain the present invention in the drawings, parts irrelevant to the description are omitted, and similar reference numerals are attached to similar parts throughout the specification.

Throughout the specification, when a part is "connected" with another part, this includes not only the case of being "directly connected" but also the case of being "electrically connected" with another element interposed therebetween. . In addition, when a part "includes" a certain component, this means that other components may be further included rather than excluding other components unless otherwise stated.

In this specification, a "part" includes a unit realized by hardware, a unit realized by software, and a unit realized using both. In addition, one unit may be implemented using two or more hardware, and two or more units may be implemented by one hardware. Meanwhile, '~ unit' is not limited to software or hardware, and '~ unit' may be configured to be in an addressable storage medium or may be configured to reproduce one or more processors. Thus, as an example, '~' denotes components such as software components, object-oriented software components, class components, and task components, and processes, functions, properties, and procedures. , subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays and variables. The functions provided in the components and '~ units' may be combined into a smaller number of components and '~ units' or further separated into additional components and '~ units'. In addition, components and '~ units' may be implemented to regenerate one or more CPUs in a device.

In the present invention, the sound signal is input to the microphone of the voice recognition service providing terminal 100, and may mean a signal for ambient noise (hereinafter referred to as an ambient noise signal) or a signal for sound including ambient noise and a voice command. can

In addition, in the present invention, the voice command signal is a signal for a voice command generated by a user uttering to use the voice recognition service, and a signal for an activation command for activating the voice recognition service (hereinafter, an activation command signal) and a signal for a control command for using the activated voice recognition service (hereinafter, a control command signal).

1 is a block diagram of a system for providing a voice recognition service according to an embodiment of the present invention.

Referring to FIG. 1 , a system for providing a voice recognition service may include a voice recognition service providing terminal 100 and a voice recognition server 110 . However, since the voice recognition service providing system of FIG. 1 is only an embodiment of the present invention, the present invention is not limitedly interpreted through FIG. 1 , and may be configured differently from FIG. 1 according to various embodiments of the present invention. have.

In general, each component of the voice recognition service providing system of FIG. 1 is connected through a network (not shown). A network refers to a connection structure that enables information exchange between each node, such as terminals and servers, and includes a local area network (LAN), a wide area network (WAN), and the Internet (WWW: World). Wide Web), wired and wireless data communication networks, telephone networks, wired and wireless television networks, and the like. Examples of wireless data communication networks include 3G, 4G, 5G, 3rd Generation Partnership Project (3GPP), Long Term Evolution (LTE), World Interoperability for Microwave Access (WIMAX), Wi-Fi, Bluetooth communication, infrared communication, ultrasound Communication, Visible Light Communication (VLC), LiFi, etc. are included, but are not limited thereto.

The voice recognition service providing terminal 100 may include all terminals providing a voice recognition service including, for example, a voice recognition speaker, a smart phone, a vehicle speaker, and the like.

The voice recognition service providing terminal 100 may receive a spoken voice from a user and perform voice recognition on the inputted spoken voice. Specifically, the voice recognition service providing terminal 100 may perform voice recognition after detecting an activation command signal (wake up word, WUW). Here, the activation command signal is a trigger voice command for activating the voice recognition function.

For example, in the voice recognition service providing terminal 100 , a first microphone 10 is installed on one side of the voice recognition service providing terminal 100 , and a second microphone is installed on the other side of the voice recognition service providing terminal 100 . 20 is installed, and a speaker 30 may be installed in a position close to any one microphone.

For example, in the voice recognition service providing terminal 100 , a plurality of microphones (not shown) may be additionally installed at positions other than the positions where each of the first microphone 10 and the second microphone 20 are installed.

For example, in a structure in which the second microphone 20 and the speaker 30 are disposed adjacently, even if the user utters a voice command including an activation command signal for voice recognition while a sound source is output through the speaker 30 Since the sound source being output through the speaker 30 is simultaneously inputted with the corresponding voice command to each of the first microphone 10 or the second microphone 20, there are many difficulties in voice recognition.

In order to solve this problem, the voice recognition service providing terminal 100 receives a sound signal through each of the first microphone 10 and the second microphone 20 , and the sound signal input through the first microphone 10 . (hereinafter, referred to as a first sound signal) and a sound signal input through the second microphone 20 (hereinafter referred to as a second sound signal), respectively, may measure sound pressure levels. Here, the sound pressure level is a physical quantity indicating the magnitude of the sound pressure, from which the sound intensity can be measured. The unit of this sound pressure level is decibel (dB).

According to an embodiment of the present invention, when a sound signal is input to each of the first microphone 10 and the second microphone 20 , the voice recognition service providing terminal 100 receives the second input through the first microphone 10 . The first sound signal and the second sound signal input through the second microphone 20 may be transmitted to the voice recognition server 110 .

At this time, the voice recognition server 110 measures the sound pressure level of the first sound signal corresponding to the first microphone 10 received from the voice recognition service providing terminal 100 , and the second microphone 20 A sound pressure level of the second sound signal may be measured.

The voice recognition service providing terminal 100 may select one of the first microphone 10 and the second microphone 20 as a microphone for receiving a voice command for receiving a voice command signal based on the measured sound pressure level.

According to an embodiment of the present invention, one of the first microphone 10 and the second microphone 20 is given a voice command based on the sound pressure level for the first sound signal and the second sound signal in the voice recognition server 110 . It can be selected as a microphone for receiving voice commands that receives a signal.

For example, when the voice recognition service providing terminal 100 measures the sound pressure level, the voice recognition service providing terminal 100 sets the sound pressure level for the first sound signal and the sound pressure level for the second sound signal to the voice recognition server. In response to transmission to 110 , the voice recognition server 110 may select a microphone for receiving a voice command based on sound pressure levels for the first sound signal and the second sound signal.

On the other hand, when the voice recognition server 110 measures the sound pressure level for the first sound signal and the second sound signal, the sound pressure for the first sound signal and the second sound signal measured by the voice recognition server 110 . A microphone for receiving voice commands may be selected based on the level.

The voice recognition service providing terminal 100 may perform voice recognition in conjunction with the voice recognition server 110 based on a voice command signal input through the selected voice command reception microphone.

At this time, when the voice recognition server 110 selects a microphone for receiving a voice command, the voice recognition service providing terminal 100 receives information about the microphone for receiving a voice command from the voice recognition server 110, and the number of corresponding voice commands Voice recognition may be performed based on a voice command signal input through the credit microphone.

For example, the voice recognition service providing terminal 100 provides a voice recognition service based on a user's activation command signal input through a microphone selected as a microphone for receiving a voice command among the first microphone 10 and the second microphone 20 . can be activated. At this time, the voice recognition performing unit 230 transmits an activation command signal to the voice recognition server 110 , and prepares for voice recognition based on voice recognition information corresponding to the activation command signal received from the voice recognition server 110 . can do.

Through this, the present invention can provide a more stable voice recognition service because the activation command signal can be easily detected even in a noisy environment such as when a sound source is being output through the speaker 30 .

Hereinafter, the operation of each component of the voice recognition service providing system of FIG. 1 will be described in more detail.

2 is a block diagram of the voice recognition service providing terminal 100 shown in FIG. 1 according to an embodiment of the present invention.

Referring to FIG. 2 , the voice recognition service providing terminal 100 according to an embodiment of the present invention includes a sound input unit 200 , a sound pressure level measurement unit 210 , a microphone selector 220 for receiving a voice command, and voice recognition. It may include a performing unit 230 and a preprocessing performing unit 240 . However, the configuration of the voice recognition service providing terminal 100 according to an embodiment of the present invention may be different from that of FIG. 1 .

The sound input unit 200 is a sound signal through each of the first microphone 10 disposed on one side of the voice recognition service providing terminal 100 and the second microphone 20 disposed on the other side of the voice recognition service providing terminal 100 . can be input. Here, the first microphone 10 may be a main microphone, and the second microphone 20 may be a sub microphone. The first microphone 10 and the second microphone 20 may be, for example, stereo microphones using an audio channel.

For example, when the user of the voice recognition service providing terminal 100 utters a voice command, the sound input unit 200 may include the user's voice command signal and the user's voice command signal through each of the first microphone 10 and the second microphone 20 . A sound signal including an ambient noise signal (eg, noise not related to a user's voice command, such as a sound source being output from the speaker 30 or a wind sound of an air conditioner) may be input.

The sound pressure level measuring unit 210 may measure the sound pressure level of each of the first sound signal input through the first microphone 10 and the second sound signal input through the second microphone 20 .

The sound pressure level measuring unit 210 measures the sound pressure level for each of the first sound signal input through the first microphone 10 and the second sound signal input through the second microphone 20 every first unit time. can Here, the first unit time at which the sound pressure level of the sound signal is measured may vary according to, for example, the specification of the voice recognition service providing terminal 100 or the voice recognition algorithm, for example, may be determined within 80ms to 120ms . Preferably, the first unit time may be 100 ms.

For example, the higher the specification for voice recognition of the voice recognition service providing terminal 100 is, the higher the voice recognition rate may be, so the first unit time may increase, and if the specification for voice recognition is low, the voice recognition rate should be increased. Therefore, the first unit time may be shortened. In this case, the first unit time may be preset or may be changed later by the user.

Here, the microphone selection unit 220 for receiving a voice command may select a microphone for receiving a voice command by comparing the sound pressure level of the first sound signal and the sound pressure level of the second sound signal measured every first unit time. In this case, the microphone selection unit 220 for receiving a voice command may store information on the microphone selected for each first unit time in a buffer of the voice recognition service providing terminal 100 .

For example, the microphone selector 220 for receiving a voice command may include a sound pressure level for the first sound signal 301 of the first microphone 10 and the second of the second microphone 20 measured every first unit time. After selecting a microphone to which a sound signal having a lower sound pressure level among sound pressure levels for the sound signal 303 is input as a microphone for receiving a voice command, information on the selected microphone may be sequentially stored in a buffer.

The microphone selection unit 220 for receiving a voice command may use information about the microphone stored in the buffer to select a microphone for receiving a voice command. For example, the microphone selection unit 220 for receiving a voice command may include a sound pressure level for a first sound signal input through the first microphone 10 and a second microphone input through the second microphone 20 , as will be described later. 2 When the sound pressure levels for the sound signals are the same or similar, a microphone for receiving a voice command may be selected using information about the microphone stored in the buffer.

Also, as described later, the microphone selection unit 220 for receiving a voice command may use information about the microphone stored in the buffer even when selecting a microphone for receiving a voice command or a microphone for receiving an additional voice command signal. Detailed information on this will be described later.

The microphone selection unit 220 for receiving a voice command may select one of the first microphone 10 and the second microphone 20 as a microphone for receiving a voice command signal for receiving a voice command signal based on the measured sound pressure level. .

For example, the microphone selection unit 220 for receiving a voice command may include a sound pressure level for a first sound signal input through the first microphone 10 and a second sound signal input through the second microphone 20 . A microphone to which a sound signal having a lower sound pressure level among sound pressure levels is input may be selected as a microphone for receiving a voice command.

According to an embodiment of the present invention, the microphone selection unit 220 for receiving a voice command is configured to receive voice commands from the first microphone 10 and the second microphone 20 for a preset past time (eg, from 10 minutes to the present). Different weights may be applied to the respective sound pressure levels of the first microphone 10 and the second microphone 20 based on the number of times the microphone is selected for receiving a command.

That is, the microphone selection unit 220 for receiving a voice command applies a first weight to the sound pressure level for a sound signal input through a microphone that has been selected as a microphone for receiving a voice command during a preset past time, and is applied through another microphone. A second weight lower than the first weight is applied to the sound pressure level of the input sound signal, and the microphone to which the sound signal having the lower sound pressure level among the sound pressure level to which the first weight is applied and the sound pressure level to which the second weight is applied is input is voiced. It can be selected as a microphone for receiving commands.

For another example, the microphone selection unit 220 for receiving a voice command may include the first microphone 10 based on the utterance position of the user's voice command (ie, the position of the user with respect to the voice recognition service providing terminal 100 ). ) and the second microphone 20, different weights may be applied to the respective sound pressure levels. To this end, the voice sound pressure level measurement unit 210 measures the sound pressure level of the voice command signal input to the first microphone 10 and the second microphone 20, and tracks the utterance position of the user's voice command based thereon. can do.

That is, the microphone selection unit 220 for receiving a voice command is a sound pressure level for a sound signal input through a microphone (the higher the sound pressure level, the closer the microphone is to the user's speaking position) located close to the user's voice command utterance position. The first weight is applied to the sound pressure level for the sound signal input through the microphone (the lower the sound pressure level, the farther away from the user's speaking position) is the microphone located far from the user's speech command location. After applying a second weight lower than the first weight, a microphone to which a sound signal having a lower sound pressure level among the sound pressure level to which the first weight is applied and the sound pressure level to which the second weight is applied is input may be selected as a microphone for receiving a voice command.

As another example, the microphone selection unit 220 for receiving a voice command is the inclination of the voice recognition service providing terminal 100 based on sensing data sensed by at least one sensor provided in the voice recognition service providing terminal 100 . Based on the information, different weights may be applied to the sound pressure levels of the first microphone 10 and the second microphone 20, respectively.

That is, the microphone selection unit 220 for receiving a voice command considers the tilt information of the voice recognition service providing terminal 100 and the sound pressure level for the sound signal input through the microphone located close to the utterance position of the user's voice command. A first weight is applied to , and a second weight lower than the first weight is applied to the sound pressure level of the sound signal input through a microphone located far from the utterance position of the user's voice command, and then the first weight is applied A microphone to which a sound signal having a lower sound pressure level is input among the sound pressure level and the sound pressure level to which the second weight is applied may be selected as a microphone for receiving a voice command.

According to the above configuration of the present invention, the performance of voice recognition can be further improved by allowing the microphone selection unit 220 for receiving a voice command to actually select a microphone for easily recognizing a user's voice command as a microphone for receiving a voice command.

According to another embodiment of the present invention, the microphone selection unit 220 for receiving a voice command is configured when the sound pressure levels of the first sound signal and the second sound signal are the same or the difference between the sound pressure levels is within a preset value, A microphone for receiving a voice command may be selected based on at least one other piece of information.

For example, the microphone selection unit 220 for receiving a voice command is configured to select the first microphone 10 and the second microphone 20 as the microphone for receiving a voice command for a preset past time (eg, from 10 minutes to the present). You can select a microphone for receiving voice commands based on the number of times. That is, the microphone selection unit 220 for receiving a voice command may select a microphone for receiving a voice command from many microphones selected as a microphone for receiving a voice command during a preset past time.

For another example, the microphone selection unit 220 for receiving a voice command may include a microphone for receiving a voice command based on the utterance position of the user's voice command (ie, the position of the user with respect to the voice recognition service providing terminal 100 ). can be selected.

That is, the microphone selection unit 220 for receiving a voice command selects a corresponding microphone (ie, a microphone located at the utterance position of the user's voice command) based on the tracked utterance position of the user's voice command. can be selected.

As another example, the microphone selection unit 220 for receiving a voice command may select a microphone for receiving a voice command based on an arrangement form of the voice recognition service providing terminal 100 and a location where a user's voice command is uttered.

To this end, the microphone selection unit 220 for receiving a voice command selects tilt information of the voice recognition service providing terminal 100 based on sensing data sensed by at least one sensor provided in the voice recognition service providing terminal 100 . can be extracted. For example, the direction in which the voice recognition service providing terminal 100 is inclined and the value of the inclination may be extracted from the angular velocity of the gyro sensor.

At this time, the microphone selection unit 220 for receiving a voice command selects a microphone for receiving a voice command based on the utterance position of the user's voice command based on the tilt information of the voice recognition service providing terminal 100 . can

As described above, according to the present invention, it is possible to select an optimal microphone capable of improving the performance of voice recognition even if the sound pressure levels of the first sound signal and the second sound signal are the same or similar. , reference numeral 305 denotes a microphone selected as a microphone for receiving a voice command every unit time of 20 ms, and reference numeral 307 denotes a microphone selected as a microphone for receiving a voice command every unit time of 100 ms.

Referring to reference numerals 305 and 307 , when a microphone for receiving a voice command is selected by measuring the sound pressure level every unit time of 20 ms, there is a problem in that the selected microphone is frequently changed.

On the other hand, when a microphone for receiving a voice command is selected by measuring the sound pressure level every 100 ms unit time, it can be confirmed that a problem in which the selected microphone is frequently changed does not occur.

However, as described above, the first unit time may vary according to the specification of the voice recognition service providing terminal 100 or the voice recognition algorithm within 80 ms to 120 ms.

According to an embodiment of the present invention, the sound pressure level measuring unit 210 compares the sound pressure level for the first sound signal and the sound pressure level for the second sound signal measured every first unit time to the first microphone 10 . and a count value may be assigned to a counter corresponding to one of the second microphones 20 .

For example, the sound pressure level measuring unit 210 may measure the sound pressure level of the first sound signal input to the first microphone 10 every first unit time for the second sound signal input to the second microphone 20 . If it is lower than the sound pressure level, a first count value (eg, +1) is assigned to the counter corresponding to the first microphone 10 , and the sound pressure level for the second sound signal input to the second microphone 20 is the second When the sound pressure level of the first sound signal input to the first microphone 10 is low, a second count value (eg, -1) may be assigned to a counter corresponding to the first microphone 10 .

The microphone selection unit 220 for receiving a voice command is based on a comparison result between the total count value of the counter corresponding to one of the first microphone 10 and the second microphone 20 and a preset threshold value, the first microphone One of the microphones (10) and the second microphone 20 may be selected as a microphone for receiving a voice command.

For example, when the total count value of the counter corresponding to the first microphone 10 exceeds a preset threshold value (eg, 0), the microphone selection unit 220 for receiving the voice command may include the first microphone 10 . can be selected as the microphone for receiving voice commands. When the total count value of the counter corresponding to the first microphone 10 is less than a preset threshold (eg, 0), the microphone selection unit 220 for receiving a voice command uses the second microphone 20 as a microphone for receiving a voice command. You can choose. Through this, it is possible to prevent the microphone selection from being frequently changed in the selection of the microphone for receiving the voice command.

Returning to FIG. 2 , the pre-processing unit 240 may perform pre-processing on the voice command signal input through the selected voice command receiving microphone. For example, the preprocessing unit 240 may perform preprocessing of inputting a sound signal input through a microphone for receiving a selected voice command to a noise filter to remove a noise noise signal from the sound signal, and extracting only the voice command signal. have.

Also, the preprocessing unit 240 may convert the voice command signal extracted from the sound signal into text, and transmit the converted text to the voice recognition performing unit 230 .

In this case, the voice recognition performing unit 230 may determine whether the text received from the preprocessing performing unit 240 is a text corresponding to the activation command signal.

The voice recognition performing unit 230 may perform voice recognition in conjunction with the voice recognition server 110 based on a voice command signal input through the selected voice command receiving microphone.

The voice recognition performer 230 may determine whether the voice command signal extracted from the sound signal input through the selected voice command reception microphone is an activation command signal for voice recognition.

For example, the voice recognition performing unit 230 may determine whether the extracted voice command signal is registered in the activation command list (a preset activation command signal or an activation command signal registered for each user ID).

In addition, the voice recognition performing unit 230 determines whether the extracted voice command signal is an activation command signal for activating the voice recognition service (eg, 'Hey Kakao') or a control command signal for using the activated voice recognition service (eg, 'Hey Kakao'). , 'tell me the weather').

When it is determined that the voice command signal is an activation command signal for voice recognition, the voice recognition performer 230 may activate the voice recognition service providing terminal 100 to receive the additional voice command.

When it is determined by the voice recognition performing unit 230 that the voice command signal is an activation command signal, the microphone selection unit 220 for receiving a voice command sends an additional voice command signal (voice command other than the activation command) to the first microphone ( 10) and at least one of the second microphone 20 may be selected as a microphone for receiving an additional voice command to receive an additional voice command signal.

For example, the microphone selection unit 220 for receiving a voice command may select a microphone receiving an activation command signal as a microphone for receiving an additional voice command receiving an additional voice command signal.

As another example, the microphone selector 220 for receiving a voice command may select the first microphone 10 set as the main microphone as a microphone for receiving an additional voice command to receive an additional voice command signal.

As another example, the microphone selection unit 220 for receiving a voice command is a voice command signal or additional voice input through each of the first microphone 10 and the second microphone 20 measured by the sound pressure level measurement unit 210 . Based on the sound pressure level for the command signal, one of the first microphone 10 and the second microphone 20 may be selected again as a microphone for receiving an additional voice command.

For example, the microphone selection unit 220 for receiving a voice command may include the first microphone ( 10 ) and the second microphone ( 20 ) based on the number of times the first microphone 10 and the second microphone 20 are selected as the microphone for receiving an additional voice command for a preset past time. 10) and the second microphone 20, a microphone for receiving an additional voice command to receive an additional voice command signal may be selected.

For example, the microphone selection unit 220 for receiving a voice command may add any one of the first microphone 10 and the second microphone 20 to receive an additional voice command signal based on the utterance position of the user's voice command. You can also choose a microphone for receiving voice commands. In this case, as described above, tilt information of the voice recognition service providing terminal 100 may be additionally taken into consideration.

Alternatively, a microphone selected as a microphone for receiving an additional voice command from which an additional voice command signal is most recently received may be equally selected.

The voice recognition performing unit 230 may include a user's additional voice command signal (eg, 'play music') input through at least one microphone selected as a microphone for receiving an additional voice command among the first microphone 10 and the second microphone 20 . '), voice recognition may be performed. At this time, the voice recognition performing unit 230 transmits an additional voice command signal to the voice recognition server 110 , and performs voice recognition based on voice recognition information corresponding to the additional voice command signal received from the voice recognition server 110 . service can be provided. Here, the voice recognition information includes response information to an additional voice command signal (eg, in the case of a question) (eg, an answer to a question), and control information (eg, on, off, volume control) of the voice recognition service providing terminal 100 . information), and control information of various terminals interworking with the voice recognition service providing terminal 100 .

4 is a flowchart illustrating a method of providing a voice recognition service according to an embodiment of the present invention. Referring to FIG. 4 , in step S401 , the voice recognition service providing terminal 100 is disposed on the other side of the first microphone 10 and the voice recognition service providing terminal 100 disposed on one side of the voice recognition service providing terminal 100 . A sound signal may be input through each of the second microphones 20 .

In step S403 , the voice recognition service providing terminal 100 may measure the sound pressure level for each of the first sound signal input through the first microphone 10 and the second sound signal input through the second microphone 20 . have.

In step S405, the voice recognition service providing terminal 100 may select one of the first microphone 10 and the second microphone 20 as a microphone for receiving a voice command for receiving a voice command signal based on the measured sound pressure level. have.

In step S407, the voice recognition service providing terminal 100 may perform voice recognition based on the voice command signal input through the selected command receiving microphone.

5 is a block diagram of the voice recognition server 110 shown in FIG. 1 according to an embodiment of the present invention.

Referring to FIG. 5 , the voice recognition server 110 according to an embodiment of the present invention includes a sound receiving unit 500 , a sound pressure level measuring unit 510 , a voice command receiving microphone selection unit 520 , and a voice recognition performing unit. 530 and a pre-processing unit 540 may be included. However, the configuration of the voice recognition server 110 according to an embodiment of the present invention may be different from that of FIG. 1 .

The sound receiver 500 receives the first sound signal input through the first microphone 10 of the voice recognition service providing terminal 100 and the second sound signal input through the second microphone 20 of the voice recognition service providing terminal It can be received from (100). Here, the first sound signal and the second sound signal are the same signal, and the first and second sound signals may include a user's voice command and an ambient noise signal.

The sound receiver 500 may receive the first sound signal and the second sound signal from the voice recognition service providing terminal 100 every first unit time. Here, the first unit time may vary according to, for example, a specification of the voice recognition service providing terminal 100 or a voice recognition algorithm, and may be determined within, for example, 80 ms to 120 ms. Preferably, the first unit time may be 100 ms.

The sound pressure level measuring unit 510 may measure the sound pressure level of each of the first sound signal and the second sound signal.

The sound pressure level measuring unit 510 measures the sound pressure level for each of the first sound signal input through the first microphone 10 and the second sound signal input through the second microphone 20 every first unit time. can

The microphone selector 520 for receiving a voice command may select one of the first microphone 10 and the second microphone 20 as a microphone for receiving a voice command for receiving a voice command signal based on the measured sound pressure level. . For example, the microphone selection unit 520 for receiving a voice command sets a microphone to which a sound signal having a lower sound pressure level among a sound pressure level for the first sound signal and a sound pressure level for the second sound signal is input as a microphone for receiving a voice command. You can choose.

The microphone selection unit 520 for receiving a voice command may select a microphone for receiving a voice command by comparing the sound pressure level of the first sound signal and the sound pressure level of the second sound signal measured every first unit time.

The microphone selection unit 520 for receiving a voice command may select a microphone for receiving a voice command through various methods in which the above-described voice recognition service providing terminal 100 selects a microphone for receiving a voice command.

When the voice command signal input through the selected voice command reception microphone is received from the voice recognition service providing terminal 110 , the preprocessing unit 540 may perform preprocessing on the voice command signal.

For example, the pre-processing unit 540 may perform pre-processing of removing a noise noise signal from a sound signal by inputting a sound signal input through a microphone for receiving a selected voice command to a noise filter, and extracting only a voice command signal. have.

Also, the preprocessing unit 540 may convert the voice command signal extracted from the sound signal into text, and transmit the converted text to the voice recognition performing unit 530 . In this case, the voice recognition performing unit 530 may determine whether the text received from the preprocessing performing unit 540 is a text corresponding to the activation command signal.

The voice recognition performer 530 may provide a voice recognition service by interworking with the voice recognition service providing terminal 100 based on a voice command signal input through the selected voice command reception microphone.

The voice recognition performing unit 530 extracts a voice command signal from a sound signal input through a microphone for receiving a selected voice command, and whether the extracted voice command signal is an activation command signal for activating a voice recognition service or a voice recognition service It may be determined whether a control command signal to be used.

When it is determined that the voice command signal is an activation command signal for voice recognition, the voice recognition performing unit 530 is configured to receive an additional voice command signal from the voice recognition service providing terminal 100 to the voice recognition service providing terminal 100 . You can pass an activation message.

When it is determined that the voice command signal is an activation command signal, the microphone selection unit 520 for receiving a voice command adds at least one microphone among the first microphone 10 and the second microphone 20 to an additional voice command signal (activation command). You can select a microphone for receiving additional voice commands to receive other voice commands).

The above-described voice recognition service providing terminal 100 may select the microphone for receiving the additional voice command by the microphone selection unit 520 for receiving the voice command through various methods for selecting the microphone for receiving the additional voice command.

When the voice recognition performing unit 530 receives the user's additional voice command signal input through the microphone for receiving the additional voice command from the voice recognition service providing terminal 100, the voice recognition is performed based on the received additional voice command signal. performed, and as a result of this, voice recognition information may be transmitted to the voice recognition service providing terminal 100 . Here, the voice recognition information includes response information to an additional voice command signal (eg, in the case of a question) (eg, an answer to a question), and control information (eg, on, off, volume control) of the voice recognition service providing terminal 100 . information), and control information of various terminals interworking with the voice recognition service providing terminal 100 .

Embodiments of the present invention may also be implemented in the form of a recording medium including instructions executable by a computer, such as a program module to be executed by a computer. Computer-readable media can be any available media that can be accessed by a computer and includes both volatile and nonvolatile media, removable and non-removable media. Computer-readable media may also include computer storage media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. The description of the present invention described above is for illustration, and those of ordinary skill in the art to which the present invention pertains can understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. For example, each component described as a single type may be implemented in a dispersed form, and likewise components described as distributed may be implemented in a combined form.

Although the methods and systems of the present invention have been described with reference to specific embodiments, some or all of their components or operations may be implemented using a computer system having a general purpose hardware architecture.

The description of the present invention described above is for illustration, and those of ordinary skill in the art to which the present invention pertains can understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. For example, each component described as a single type may be implemented in a dispersed form, and likewise components described as distributed may be implemented in a combined form.

The scope of the present invention is indicated by the following claims rather than the above detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention. do.

Claims (18)

In the terminal providing a voice recognition service,
a sound input unit receiving a sound signal through each of a first microphone disposed on one side of the terminal and a second microphone disposed on the other side of the terminal;
a sound pressure level measuring unit for measuring sound pressure levels for each of the first sound signal input through the first microphone and the second sound signal input through the second microphone;
a microphone selection unit for receiving a voice command for selecting one of the first microphone and the second microphone as a microphone for receiving a voice command for receiving a voice command signal based on the measured sound pressure level; and
A voice recognition performing unit for performing voice recognition based on the voice command signal input through the selected command receiving microphone
including,
The microphone selection unit for receiving a voice command is based on the number of times that one of the first microphone and the second microphone is selected as the microphone for receiving the voice command during a preset past time, the sound pressure level of each of the first microphone and the second microphone different weights are applied to
Of the sound pressure levels to which the different weights are applied, a microphone to which a sound signal having a lower sound pressure level is input is selected as a microphone for receiving a voice command.
delete The method of claim 1,
The sound pressure level measuring unit measures the sound pressure level of the sound signal every first unit time,
The microphone selection unit for receiving a voice command compares the sound pressure level for the first sound signal and the sound pressure level for the second sound signal measured every first unit time to select the microphone for receiving the voice command, terminal.
delete The method of claim 1,
The sound pressure level measuring unit measures a sound pressure level of a first voice command signal input to the first microphone and a sound pressure level of a second voice command signal input into the second microphone, and measures the sound pressure level of the first voice command signal and the second voice command signal. 2 The terminal is to track the utterance position of the user's voice command based on the sound pressure level of the voice command signal.
6. The method of claim 5,
The microphone selection unit for receiving a voice command applies a first weight to the first sound signal and a second weight to the second sound signal based on the utterance position of the user's voice command.
delete 6. The method of claim 5,
The terminal for selecting the microphone for receiving the voice command is to select the microphone for receiving the voice command based on the utterance position of the user's voice command.
The method of claim 1,
When the voice command signal is an activation command signal for activating the voice recognition service, the voice command receiving microphone selector receives an additional voice command signal through at least one of the first microphone and the second microphone. A terminal that is selected as a microphone for receiving commands.
10. The method of claim 9,
When the voice command signal is an activation command signal for activating the voice recognition service, the microphone selection unit for receiving a voice command is a microphone for receiving the additional voice command for the first microphone and the second microphone for a preset past time. The terminal for selecting the microphone for receiving the additional voice voice command based on the selected number of times.
The method of claim 1,
The terminal further comprising a pre-processing unit for performing pre-processing on the voice command signal input through the selected command receiving microphone.
A method of providing a voice recognition service, comprising:
receiving a sound signal through each of a first microphone disposed on one side of the terminal and a second microphone disposed on the other side of the terminal;
measuring a sound pressure level for each of a first sound signal input through the first microphone and a second sound signal input through the second microphone;
selecting one of the first microphone and the second microphone as a microphone for receiving a voice command for receiving a voice command signal based on the measured sound pressure level; and
performing voice recognition based on the voice command signal input through the selected command receiving microphone
including,
The selecting step
For a preset past time, based on the number of times that one of the first microphone and the second microphone is selected as the microphone for receiving the voice command, different weights are applied to the sound pressure levels of the first microphone and the second microphone, respectively step; and
and selecting a microphone to which a sound signal having a lower sound pressure level is input from among the sound pressure levels to which the different weights are applied as a microphone for receiving a voice command.
delete 13. The method of claim 12,
Measuring the sound pressure level
Measuring the sound pressure level for the sound signal every first unit time,
The selecting step
Comparing the sound pressure level for the first sound signal and the sound pressure level for the second sound signal measured for each first unit time, selecting a microphone for receiving the voice command, the voice recognition service providing Way.
13. The method of claim 12,
and performing pre-processing on the voice command signal input through the selected command receiving microphone.
A server providing a voice recognition service, comprising:
a sound receiver for receiving sound signals input from the terminal through each of the first microphone and the second microphone of the terminal;
a sound pressure level measuring unit for measuring sound pressure levels for each of the first sound signal input through the first microphone and the second sound signal input through the second microphone;
a microphone selection unit for receiving a voice command for selecting one of the first microphone and the second microphone as a microphone for receiving a voice command for receiving a voice command signal based on the measured sound pressure level; and
A voice recognition performing unit for performing voice recognition based on the voice command signal input through the selected command receiving microphone
including,
The microphone selection unit for receiving a voice command is based on the number of times that one of the first microphone and the second microphone is selected as the microphone for receiving the voice command during a preset past time, the sound pressure level of each of the first microphone and the second microphone different weights are applied to
Of the sound pressure levels to which the different weights are applied, a microphone to which a sound signal having a lower sound pressure level is input is selected as a microphone for receiving a voice command.
17. The method of claim 16,
The voice recognition performing unit determines whether the voice command signal is an activation command signal for activating the voice recognition service, and when it is determined that the voice command signal is the activation command signal, the terminal sends an additional voice command signal A server that sends a message that activates it to be received.
18. The method of claim 17,
The voice recognition performing unit receives the additional voice command signal from the terminal, performs voice recognition based on the received additional voice command signal, and transmits voice recognition information to the terminal as a result of the voice recognition , server.

Images