JP2019086643A - Voice recognition system - Google Patents

Abstract

To provide a "voice recognition system" which always receives an urgent voice input for a specific function, while selectively using plural pieces of voice recognition means used for voice input for different functions.SOLUTION: Normally, when a voice recognition device 102 performs continuous voice recognition and inputs a command of voice recognition result to a data processing device 109 and an operation of a talk switch 107 occurs, voice recognition is started using a voice recognition server by a portable device 2. While the voice recognition by the portable device 2 is being performed, basically, the output of the command of the voice recognition result by the voice recognition device 102 to the data processing device 109 is stopped, but when the recognition result by the voice recognition device 102 is a command requesting execution of an urgent process registered in a priority word table 105, the command recognized by the voice recognition device 102 is output to the data processing device 109.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a technology of voice recognition that recognizes a user's uttered voice.

  As a speech recognition technology for recognizing a user's speech, there is known a technology for always recognizing a user's speech without requiring the user's speech recognition start instruction operation (for example, Patent Document 1). .

  In addition, as a technology for voice recognition that recognizes the user's voice, the terminal performs voice recognition of the user's voice by a voice recognition device provided by the user, and the user's voice using an external voice recognition server There is known a technology for performing speech recognition and using the speech recognition result obtained by any one of the speech recognition (for example, Patent Documents 2 and 3).

International Publication No. 2014/0687888 Unexamined-Japanese-Patent No. 2015-102795 JP 2013-064777

  In the terminal, when speech recognition of the user's speech by the speech recognition device provided by itself and speech recognition of the user's speech using an external speech recognition server are used for command input of different functions, It is not appropriate to perform both voice recognition because a command input by the user through speech is likely to be input as a command for a function different from the function intended by the user.

  Therefore, it is desirable to selectively switch between speech recognition of the user's speech by the speech recognition apparatus provided by the user and speech recognition of the user's speech using an external speech recognition server.

  On the other hand, in the case where voice recognition by the voice recognition device provided by itself is always performed without requiring the user's voice recognition start instruction operation, voice recognition of the user's uttered voice by the voice recognition device provided by itself is performed. If the voice recognition of the user's uttered voice using the external voice recognition server is selectively switched and performed, the following problems occur.

  That is, in this case, the user using only the external voice recognition server and the first mode in which the voice recognition by its own voice recognition device is always performed without the need for the user's voice recognition start instruction operation. A second mode for speech recognition of uttered speech is provided. Usually, speech recognition is performed in the first mode, and speech recognition is temporarily performed in the second mode according to the user's operation, and speech recognition is performed. It is possible to return to the first mode if you have finished, but in this way, during the second mode, a function that uses speech recognition by the speech recognition device that it has for input of commands. Even if it is necessary to input an urgently required command, the second mode is forcibly switched to the first mode because there is no speech recognition start instruction operation for speech recognition in the first mode. Can not be replaced, input of commands that require the emergency becomes impossible.

  On the other hand, even during the second mode, assuming that voice recognition is performed by the voice recognition device provided by the user, as described above, the command input by the user by the uttered voice causes the user to input the command. It may be input as a command for a function different from the intended function.

  Therefore, the present invention is a voice recognition system using the first voice recognition means that always performs voice recognition without requiring a user's voice recognition start instruction operation and the second voice recognition means for input to different functions. It is an object of the present invention to make it possible to always perform an urgent input using voice recognition of the first voice recognition means while suppressing erroneous input to each function.

  In order to achieve the above task, the present invention provides a voice recognition system for voice recognition of voice uttered by a user, the first voice recognition for performing continuous voice recognition of a microphone and a command represented by the voice collected by the microphone Means, a first functional unit that executes processing for instructing the execution of the input command, second voice recognition means for performing voice recognition operation for voice recognition of voice collected by the microphone, and the second voice recognition means A second functional unit for processing a speech recognition result of the speech recognition device; a priority command storage means in which a part of the commands of the speech recognition means is recognized as a priority command; and the speech recognition It is provided with operation control means. Here, the voice recognition operation control means controls switching between the first voice recognition mode and the second voice recognition mode provided as the voice recognition mode, and the voice recognition mode is the first voice. When in the recognition mode, the voice recognition operation of the second voice recognition means is stopped, and a command recognized by the first voice recognition means is inputted to the first functional unit, and the voice recognition mode is When the second voice recognition mode is in the second voice recognition mode, the second voice recognition means executes the voice recognition operation, and a command recognized by the first voice recognition means is stored in the priority command storage means. Only in the case of the registered priority command, the first voice recognition means inputs the voice-recognized command to the first functional unit.

  Here, in the voice recognition system, when the voice recognition operation control unit generates a predetermined input from the user when the voice recognition mode is in the first voice recognition mode, the voice recognition mode is set to the second voice recognition mode. Switching to the second voice recognition mode, and when the voice recognition operation of the second voice recognition unit is completed when the second voice recognition mode is in the first voice recognition mode. It may be configured to switch to

  In addition, such a speech recognition system is connected to an external speech recognition server providing speech recognition service via the communication as the speech recognition operation, and the second speech recognition unit is connected to the speech recognition server. The voice recognition service may be used to perform voice recognition of voice collected by the microphone.

In addition, in such a speech recognition system, the speech recognition system may be a speech recognition system used for speech input in a system mounted on a car.
Further, such a voice recognition system is constituted by an on-vehicle system mounted on a vehicle and a portable device selectively connected to the on-vehicle system, and the on-vehicle system comprises the microphone and the first voice recognition. Means, the first function unit, the priority command storage means, and the voice recognition operation control means, and the portable device is provided with the second voice recognition means and the second function portion; In the operation control means, when the voice recognition mode is in the second voice recognition mode, the voice collected by the microphone is transferred from the in-vehicle system to the portable device, and in the second voice recognition means, The system may be configured to perform voice recognition operation for voice recognition of voice transferred from the on-vehicle system to the portable device.

  Further, in each of the voice recognition systems described above, among the commands in which the first voice recognition means recognizes the voice of the priority command storage means, the process in which the command instructs execution of the command relates to securing the safety of the vehicle. The command that is may be registered at least as a priority command.

  In each of the speech recognition systems described above, the priority command storage means is a significant process for the speech recognition result obtained by speech recognition of speech representing the command among the speech recognition means of the first speech recognition means. A command not defined in the second function unit at the time may be registered at least as a priority command.

  According to the speech recognition system as described above, when the first speech recognition mode performs command input to the first functional unit by the first speech recognition means, the speech recognition by the two speech recognition means is When stopping and at the same time in the second voice recognition mode and performing voice input to the second functional unit by the second voice recognition means, basically, the first voice recognition for performing constant voice recognition Stopping the input of the command to the first function unit using the speech recognition of the means;

  Therefore, a voice uttered for the user to input a command to the first functional unit may be voice-inputted to the second functional unit, or a voice uttered for the user to voice-input the second functional unit. Is erroneously recognized as the voice of the command input to the first functional unit, and erroneous input of the command to the first functional unit is suppressed.

  On the other hand, even when the second voice recognition means performs voice input to the second functional unit, the priority command registered in the priority command storage means is recognized by the voice recognition of the first voice recognition means. If this is the case, this is input to the first function unit.

  Therefore, the first command is always registered in the priority command storage means as a priority command in which the command instructing the execution of the urgent processing such as the command instructing the execution of the processing relating to securing the safety of the vehicle is registered as the priority command. It becomes possible to input to the first function unit a command instructing execution of an urgent process using the speech recognition of the speech recognition means.

  In addition, the command of the first functional unit whose significant processing with respect to the speech recognition result of speech recognition of the voice representing the command is not defined in the second functional unit at the time, that is, the command to the second functional unit The voice recognition of the first voice recognition means is always used for these commands by registering in the priority command storage means the command having no fault even if the voice representing the voice has been input as the priority command. The command can be input to the first function unit.

  As described above, according to the present invention, the first voice recognition means and the second voice recognition means, which always perform voice recognition without requiring the user's voice recognition start instruction operation, are input to different functions. In the speech recognition system used in the present invention, it is possible to always perform an urgent input using speech recognition of the first speech recognition means while suppressing erroneous input to each function.

It is a block diagram showing composition of an information processing system concerning an embodiment of the present invention. It is a flowchart which shows the external speech recognition control processing which concerns on embodiment of this invention. It is a flowchart which shows the speech recognition result filter process which concerns on embodiment of this invention. It is a sequence diagram which shows the operation example of the information processing system which concerns on embodiment of this invention.

Hereinafter, an embodiment of the present invention will be described by taking an application to an information processing system used in a car as an example.
FIG. 1 shows the configuration of the information processing system.
As illustrated, the information processing system includes an in-vehicle system 1 mounted in a car and a portable device 2 selectively connected to the in-vehicle system 1.
Here, the portable device 2 is, for example, a device portable by a user such as a smartphone or a tablet device. In addition, the portable device 2 is connected to the external voice recognition server 3 via mobile communication, and performs voice recognition of the voice transferred from the in-vehicle system 1 using the voice recognition service of the voice recognition server 3 for voice recognition Is accepted as an audio input to the portable device 2 and has a function of performing an operation according to the audio input.

  Next, the in-vehicle system 1 communicates with the microphone 101, the voice recognition device 102, the voice recognition dictionary 103, the voice recognition result filter unit 104, the priority word table 105, the external voice recognition control unit 106, the talk switch 107, and the portable device 2. A communication interface 108 to perform, a data processing device 109, and various peripheral devices 110 such as a display, a camera for photographing the surroundings of the vehicle, an AV device, an air conditioner, and the like are provided.

  Then, in the voice recognition dictionary 103, data for voice recognition of a word representing a command of the data processing device 109 is registered. Then, the voice recognition device 102 uses the voice recognition dictionary 103 to recognize the command when the user's uttered voice input from the microphone 101 is a voice representing a command of the data processing device 109, as a voice recognition result. The operation of outputting to the speech recognition result filter unit 104 is always performed without using the speech recognition start instruction operation of the user as a trigger.

  Next, in the priority word table 105, a command that needs to be urgently processed among the commands represented by the words for which the data for voice recognition is registered in the voice recognition dictionary 103 and the voice input to the portable device 2 A command that never appears is registered. Then, the speech recognition result filter unit 104 performs speech recognition result filtering processing, which will be described in detail later, using the priority word table 105 to satisfy a predetermined condition among the speech recognition results recognized by the speech recognition apparatus 102. Only the speech recognition result is output to the data processor 109.

  The command that needs to be processed urgently is, for example, a command that instructs the data processing apparatus 109 to execute the process related to securing the safety of the vehicle. For example, the data processing apparatus 109 captures the back of the vehicle A command "back camera" or the like that instructs display of an image captured by a camera can be a command that needs to be urgently processed.

  Further, as a command that does not appear in the voice input to the portable device 2, for example, when the portable device 2 uses voice input for inputting a command to the portable device 2 at that time, the voice recognition dictionary 103 Among the commands represented by the word in which the data for recognition is registered, the command may not correspond to the portable device 2.

  Further, among commands represented by words for which data for voice recognition is registered in the voice recognition dictionary 103, commands that are unlikely to appear in the voice input to the portable device 2 are also voice input to the portable device 2 It may be a command that never appears in.

  Here, such a command that does not appear in the voice input to the portable device 2 includes, for example, a command “air volume up” that instructs the data processing device 109 to execute the process of increasing the air volume of the air conditioner.

Then, the data processing device 109 performs processing according to the command output from the speech recognition result filter unit 104.
Also, the external voice recognition control unit 106 performs external voice recognition control processing, which will be described in detail later, and transfers the user's uttered voice input from the microphone 101 to the portable device 2 via the communication interface 108. The voice recognition using the voice recognition service of the voice recognition server 3 described above is performed.

The external speech recognition control process performed by the external speech recognition control unit 106 will be described below.
FIG. 2 shows the procedure of the external speech recognition control process.
As shown, in the external speech recognition control process, the external speech recognition control unit 106 monitors that the talk switch 107 is turned on by the user (step 202), and the external speech recognition mode is set if it is turned on. It sets (step 204).

  Then, a voice recognition start command is issued to the portable device 2 through the communication interface 108 (step 206), and transfer to the portable device 2 through the communication interface 108 of the user's speech input from the microphone 101 is started ( Step 208).

  Here, when the portable device 2 receives the voice recognition start command from the in-vehicle system 1, the portable device 2 transfers a series of voices transferred from the in-vehicle system 1 using the start point of the silent section longer than the predetermined time as an end point The voice recognition process for performing voice recognition using the voice recognition service of the voice recognition server 3 is started.

  Next, when the external speech recognition control unit 106 starts transfer of the user's speech to the portable device 2 through the communication interface 108 (step 208), the external speech recognition from the speech recognition result filter unit 104 is stopped. The occurrence of reception of a command (step 210) and occurrence of notification of completion of voice recognition from the portable device 2 via the communication interface 108 (step 212) are monitored.

Here, when the voice recognition process described above is completed, the portable device 2 outputs a voice recognition end notification to the on-vehicle device.
Then, the external speech recognition control unit 106 either receives an external speech recognition stop command from the speech recognition result filter unit 104 (step 210) or receives a speech recognition end notification from the portable device 2 (step 212). If it occurs, the transfer to the portable device 2 through the communication interface 108 of the user's speech input from the microphone 101 is stopped (step 214), and the external speech recognition mode is released (step 216).

Then, the process returns to the process from step 202.
The external speech recognition control process performed by the external speech recognition control unit 106 has been described above.
Next, the above-described speech recognition result filtering process performed by the speech recognition result filter unit 104 will be described.
FIG. 3 shows the procedure of the speech recognition result filtering process.
As shown, the voice recognition result filter unit 104 waits for the input of the voice recognition result from the voice recognition device 102 in the voice recognition result filter processing (step 302), and if the voice recognition result is input, the external voice recognition control is performed. It is checked whether the external speech recognition mode is set by the unit 106 (step 304).

  Then, if the external speech recognition mode is not set (step 304), the input speech recognition result is output to the data processing apparatus 109 (step 310), and the process returns to the process from step 302.

  On the other hand, when the external speech recognition mode is set (step 304), it is checked whether the input speech recognition result is a command registered in the priority word table 105 (step 306).

  If the input speech recognition result is not a command registered in the priority word table 105 (step 306), the received speech recognition result is discarded, and the process returns to step 302.

  On the other hand, if the received voice recognition result is a command registered in the priority word table 105 (step 306), an external voice recognition stop command is sent to the external voice recognition control unit 106 (step 308), The received speech recognition result is output to the data processor 109 (step 310), and the process returns to step 302.

The speech recognition result filtering process performed by the speech recognition result filter unit 104 has been described above.
Here, FIG. 4 shows an example of the operation of speech recognition by the external speech recognition control process and the speech recognition result filtering process as described above.
As shown in the figure, usually, the voice uttered by the user input from the microphone 101 is sent to the voice recognition device 102 (401), voice recognition is performed in the voice recognition device 102, and the voice recognition result is sent to the voice recognition result filter unit 104. Sent (402). Then, the speech recognition result filter unit 104 outputs the received speech recognition result to the data processing device 109 (403).

  On the other hand, when the user performs the on operation of the talk switch 107 to perform voice input to the portable device 2 (411), the external voice recognition control unit 106 sets the external voice recognition mode (412).

  After that, when the user utters a word that is not a word representing a command registered in the priority word table 105, the voice (non-registered word voice) is transmitted from the microphone 101 to the voice recognition device 102 and the external voice recognition control unit 106. Sent to (413).

  The speech recognition device 102 performs speech recognition on the received speech, and sends the speech recognition result to the speech recognition result filter unit 104 (414). Since the received speech recognition result is not a command registered in the priority word table 105, the speech recognition result filter unit 104 discards the speech recognition result without outputting it to the data processing device 109.

On the other hand, the external voice recognition control unit 106 transfers the received voice to the portable device 2 (415), and the portable device 2 transmits the transferred voice to the voice recognition server 3 (416).
Then, when the user speaks a word representing a command registered in the priority word table 105 before the portable device 2 receives the voice recognition result of the transmitted (416) voice from the voice recognition server 3, the voice (see FIG. The registered word voice) is sent from the microphone 101 to the voice recognition device 102 and the external voice recognition control unit 106 (417).

  The speech recognition device 102 performs speech recognition on the received speech, and sends the speech recognition result to the speech recognition result filter unit 104 (418). The voice recognition result filter unit 104 issues an external voice recognition stop command to the external voice recognition control unit 106 (419) because the received voice recognition result is a command registered in the priority word table 105 (419). Are output to the data processor 109 (420).

  On the other hand, the external voice recognition control unit 106 transfers the received voice to the portable device 2 (421). Here, it is assumed that the portable device 2 is configured to ignore the transferred voice during a period waiting for the voice recognition result from the voice recognition server 3, and the portable device 2 is transferred to the portable device 2 ( 421) The voice is discarded without being transmitted to the voice recognition server 3.

When the external speech recognition control unit 106 receives the external speech recognition stop command (419), it cancels the external speech recognition mode (422).
On the other hand, when the voice recognition result of the voice transmitted by the portable device 2 (416) is answered from the voice recognition server 3 to the portable device 2 (423), the portable device 2 processes the voice recognition result.

In addition, the voice recognition process of the portable device 2 ends, and the portable voice device 2 notifies the external voice recognition control unit 106 of the completion of voice recognition (424).
The example of the operation of the speech recognition by the external speech recognition control process and the speech recognition result filtering process has been described above.
Here, unlike the example shown in FIG. 4, immediately after the user turns on the talk switch 107, or while the user turns on the talk switch 107, the word which is not a word representing a command registered in the priority word table 105 is interrupted. Even when a word representing a command registered in the priority word table 105 is uttered after uttering, the recognition result of the speech recognition of the utterance of the word by the speech recognition device 102 is processed through the speech recognition result filter unit 104. It is output to the device 109.
In these cases, the speech of the word representing the command registered in the priority word table 105 is transferred from the external speech recognition control unit 106 to the portable device 2 and the portable device 2 uses the speech recognition server 3 for the voice. Although there may be a disadvantage that voice recognition is performed, if the voice uttered by the user represents a command that needs to be urgently processed, the input to the data processing device 109 of the command is prioritized over the inconvenience. If the voice uttered by the user represents a command that does not appear in the voice input to the portable device 2, the portable device 2 does not perform an unexpected operation based on the voice recognition result.

The embodiments of the present invention have been described above.
According to the present embodiment, normally, in a state where the speech input of the user's speech using the speech recognition server 3 by the portable device 2 is stopped, the speech recognition and speech of the regular speech of the user by the speech recognition device 102 are Although the command of the recognition result is input to the data processing device 109, when the user performs the on operation of the talk switch 107 to perform voice input to the portable device 2, the voice recognition server 3 by the portable device 2 is used. Voice input of the user's uttered voice is started.

  Then, basically, while the speech input of the user's speech using the speech recognition server 3 by the portable device 2 is being performed, the command of the speech recognition result of the speech of the user by the speech recognition device 102 is basically The input to the data processor 109 is stopped.

  Therefore, the user can input into the portable device 2 a voice uttered for inputting a command into the data processing device 109, or the user can utter a voice uttered for inputting into the portable device 2 a data processing device The erroneous recognition of the voice of the command input to 109 and the erroneous input of the command to the data processing device 109 are suppressed.

  On the other hand, even during a period in which the user's voice input using the voice recognition server 3 by the portable device 2 is performed while speech input is being performed, execution of a process requiring emergency registered in the priority word table 105 is requested. The command recognized by the voice recognition device 102 is input to the data processing device 109 only when the user utters a word representing a command to be performed or a command that does not appear in the voice input to the portable device 2.

  Therefore, according to the present embodiment, while suppressing erroneous input of a command to the data processing apparatus 109 and erroneous audio input to the portable apparatus 2, an audio of a command requesting execution of an urgent process to the data processing apparatus 109 at all times. You will be able to do input. In addition, even for commands that do not appear in voice input to the portable device 2, voice input to the data processing device 109 can always be performed.

  In the above embodiment, the on operation of the talk switch 107 is a trigger for causing the portable device 2 to perform voice recognition. However, this trigger may be other than the on operation of the talk switch 107. That is, this trigger may be voice input of a command instructing start of voice input to the portable device 2 or the like. In this case, occurrence of voice input of a command instructing start of voice input to portable device 2 is detected by voice recognition of a word representing the command in the voice of the user in voice recognition device 102. .

  Further, in the above embodiment, instead of the portable device 2 that performs speech recognition using the speech recognition server 3 for speech input from the microphone 101, speech using the speech recognition server 3 for speech input from the microphone 101 is used. Even when the portable device 2 is not a device that performs recognition, or any device that performs voice recognition of an uttered voice input from the microphone 101 using the voice recognition function provided by itself without using the voice recognition server 3. The present invention can be similarly applied by replacing the portable device 2 with the provided device.

  Reference Signs List 1 in-vehicle system 2 portable device 3 voice recognition server 101 microphone 102 voice recognition device 103 voice recognition dictionary 104 voice recognition result filter unit 105 priority word table 106 external voice Recognition control unit 107: Talk switch 108: Communication interface 109: Data processing device 110: Peripheral device.

Claims (7)

A speech recognition system for speech recognition of speech uttered by a user, comprising:
A microphone,
First voice recognition means for performing continuous voice recognition of a command represented by voice collected by the microphone;
A first functional unit that executes a process in which an input command instructs execution;
Second voice recognition means for performing voice recognition operation for voice recognition of voice collected by the microphone;
A second functional unit that processes the speech recognition result of the second speech recognition device;
Priority command storage means in which a part of voice recognition commands of the first voice recognition means is registered as a priority command;
And voice recognition operation control means,
The voice recognition operation control means
Controlling switching between a first speech recognition mode provided as a speech recognition mode and a second speech recognition mode;
When the speech recognition mode is the first speech recognition mode, the speech recognition operation of the second speech recognition means is stopped, and a command recognized by the first speech recognition means is said first speech recognition mode. Input to the function section,
When the speech recognition mode is the second speech recognition mode, the second speech recognition means is caused to execute the speech recognition operation, and a command recognized by the first speech recognition means is said priority A voice recognition system, wherein a command recognized by the first voice recognition means is inputted to the first functional unit only when the command is a priority command registered in the command storage means. The speech recognition system according to claim 1, wherein
The voice recognition operation control means switches the voice recognition mode to the second voice recognition mode when a predetermined input from the user is generated when in the first voice recognition mode. A voice recognition mode, and when the voice recognition operation of the second voice recognition unit is completed, the voice recognition mode is switched to the first voice recognition mode. The speech recognition system according to claim 1 or 2, wherein
The second speech recognition means is connected to an external speech recognition server providing speech recognition service via communication as the speech recognition operation, and uses the speech recognition service of the connected speech recognition server to transmit the microphone. A voice recognition system characterized by performing voice recognition of voices picked up in the above. The speech recognition system according to claim 1, 2 or 3, wherein
A speech recognition system characterized in that the speech recognition system is a speech recognition system used for speech input in a system mounted on a car. The speech recognition system according to claim 1, 2 or 3, wherein
An on-vehicle system mounted on a vehicle and a portable device selectively connected to the on-vehicle system;
The in-vehicle system includes the microphone, the first voice recognition means, the first functional unit, the priority command storage means, and the voice recognition operation control means. The portable device is configured to receive the second voice recognition. Means and the second functional unit,
The voice recognition operation control means transfers voice collected by the microphone from the in-vehicle system to the portable device when the voice recognition mode is in the second voice recognition mode.
The voice recognition system according to claim 1, wherein the second voice recognition means performs a voice recognition operation for voice recognition of voice transferred from the in-vehicle system to the portable device. The speech recognition system according to claim 1, 2, 3, 4 or 5, wherein
Among the commands recognized by the first voice recognition means, the priority command storage means includes at least a priority command of which the process in which the command instructs execution is a process related to securing the safety of the vehicle. A speech recognition system characterized by being registered as. The speech recognition system according to claim 1, 2, 3, 4, 5 or 6.
In the priority command storage means, significant processing is performed on the voice recognition result obtained by speech recognition of the speech representing the command among the speech recognition commands of the first speech recognition means at the time point, the second function unit A voice recognition system characterized in that at least a command not defined in is registered as a priority command.