JP2004354942A - Voice interactive system, voice interactive method and voice interactive program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a voice interactive system in which a user is able to select an input that is convenient for the user. <P>SOLUTION: The voice interactive system is provided with a talking button switch which is only pushed during a speaker is talking and the speaker controls devices while the speaker interacts with the system by voice. The system is provided with a time monitoring means which measures a lapse of time from the time when the talking button switch is pushed to the time when the switch is released, an operation setting storage means which beforehand stores operations to be performed next for every lapse of time and a control means which reads the operations corresponding to the lapse of time notified by the time monitoring means from the operation setting storage means and executes the read operations. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a voice interaction system, a voice interaction method, and a voice interaction program for acquiring a user's request by voice interaction when performing a device control process or an information providing process.
[0002]
[Prior art]
In order to reduce the influence on the driving operation of the driver in the operation of the car navigation system or the like in the vehicle, it is desirable to reduce the load of the operation in which the driver interacts with the system. On the other hand, in order to improve the recognition rate of voice recognition, the voice recognition process can be simplified if the speaker speaks while operating a switch or the like. For example, a technology is known in which a speech button is pushed only while a speaker is speaking, and the speech uttered during the time when the speech button was pushed is targeted for speech recognition (called press talk). ing. Also, by providing a cancel button in addition to the utterance button, in addition to the technology and voice operation that made it easy to return to the previous state in the case of erroneous recognition, it is also displayed on the monitor screen. There is also known a technique in which an option is indicated, and input contents are selected from the options and determined by a touch operation or a remote control operation.
[0003]
By the way, in order to reduce the driver load and the manufacturing cost at the time of dialogue, it is better to use fewer buttons and switches for the dialogue, and efforts are being made to reduce the number of buttons and switches used for other than the driving operation. For example, in an operation switch provided separately from an utterance button, a plurality of operations of short-time button push / long-time button push are realized by one operation switch (Patent Document 1).
Also, after the user inputs a voice command, the system checks with the user whether the recognition result is correct, and then executes processing corresponding to the voice command. In addition, the input can be made by a "permission switch" on the steering wheel (Patent Document 2). However, this “permission switch” has one meaning, and is independent of the speech recognition function regardless of the speech button.
Also, the utterance button and the operation button are integrated, and the system limits the use of the button to one or the other depending on the scene, such as "Speech button in this scene, operation button in this scene", and the utterance button In a specific scene, it is made to function as an operation switch (Patent Document 3).
[0004]
[Patent Document 1]
JP 2001-154689 A
[Patent Document 2]
Japanese Patent Publication No. 2002-12100
[Patent Document 3]
JP 2001-216130 A
[0005]
[Problems to be solved by the invention]
However, in the techniques of Patent Literatures 1 and 2, since the utterance button and the operation button are independently provided, it is necessary to use a plurality of buttons properly, which involves movement of a hand or finger, and movement of a line of sight. There is a problem that the load increases. Further, the technology of Patent Document 3 integrates the utterance button and the operation button into one, but is limited to only one of the functions depending on the scene. "And" Scenes to be input as operation buttons "are determined by the convenience of the system, and operations such as using the utterance button as a shortcut button equivalent to" Next "in the case of voice input at the request of the user There is a problem that can not be done.
[0006]
The present invention has been made in view of such circumstances, and it is possible to simultaneously realize a press talk function for voice input and an operation button function for realizing a shortcut of an interactive operation with a single utterance button. It is an object of the present invention to provide a voice interaction system, a voice interaction method, and a voice interaction program that enable a user to select an input that is convenient for him / her.
[0007]
[Means for Solving the Problems]
The invention according to claim 1 is a voice interaction system including an utterance button switch that is pushed only while the speaker is speaking, wherein the speaker controls the device while interacting with the system by voice, Time monitoring means for measuring an elapsed time from when the utterance button switch is pushed to when the utterance button switch is released, an operation setting storage means in which an operation to be performed next for each of the elapsed time is stored in advance, and the time monitoring means notifies Control means for reading an operation corresponding to the elapsed time from the operation setting storage means and executing the read operation.
[0008]
According to a second aspect of the present invention, the operation setting storage means includes: a first operation to be performed when the elapsed time is less than a first threshold; A second operation to be performed when the elapsed time is less than a second threshold value and a third operation to be performed when the elapsed time exceeds the second threshold value are stored in advance. The utterance button switch is an operation that functions as an utterance button, and the first operation and the third operation are operations in which the utterance button switches function as different operation buttons.
[0009]
In the invention described in claim 3, the action setting storage means stores an action for each scene when a dialogue is performed, and the first threshold value may be uttered in a target scene. It is a time shorter than the time required to utter the shortest response content of the response content, and the second threshold value is the most time responsive among the response content likely to be uttered in the target scene. It is characterized in that the time is longer than a time required for uttering a long response content.
[0010]
The invention according to claim 4 is characterized in that the voice interaction system includes input means for arbitrarily setting the first and second thresholds.
[0011]
According to a fifth aspect of the present invention, there is provided a speech dialogue method in a speech dialogue system including a speech button switch for pushing only while a speaker is speaking, and controlling a device while the speaker interacts with the system by voice. A time monitoring step of measuring an elapsed time from when the utterance button switch is pushed to when the utterance button switch is released; an operation setting storage step in which an operation to be performed next is stored in advance for each elapsed time; A control step of reading and executing an operation corresponding to the elapsed time notified by the time monitoring step from the operations stored in the operation setting storage step.
[0012]
The invention according to claim 6, wherein the operation setting storing step includes a first operation to be performed when the elapsed time is less than a first threshold, and a first operation to be performed when the elapsed time is equal to or more than the first threshold. A second operation to be performed when the elapsed time is less than a second threshold value, and a third operation to be performed when the elapsed time exceeds the second threshold value. The utterance button switch is an operation that functions as an utterance button, and the first operation and the third operation are operations in which the utterance button switches function as different operation buttons.
[0013]
In the invention described in claim 7, in the action setting storage step, an action is stored for each scene when a dialogue is performed, and the first threshold value may be uttered in a target scene. It is a time shorter than the time required to utter the shortest response content of the response content, and the second threshold value is the most time responsive among the response content likely to be uttered in the target scene. It is characterized in that the time is longer than a time required for uttering a long response content.
[0014]
The invention according to claim 8 is characterized in that the voice interaction method has an input step of arbitrarily setting the first and second thresholds.
[0015]
According to a ninth aspect of the present invention, there is provided a speech dialogue system including a speech button switch for pushing only while the speaker is speaking, and operating in a speech dialogue system in which the speaker controls a device while interacting with the system by voice. A program, a time monitoring process for measuring an elapsed time from when the utterance button switch is pushed to when the utterance button switch is released, and an operation setting storage process for previously storing an operation to be performed next for each of the elapsed times. And causing the computer to perform an operation corresponding to the elapsed time notified by the time monitoring process from among the operations stored in the operation setting storage process and executing the control process.
[0016]
In the invention according to claim 10, the operation setting storage processing includes a first operation to be performed when the elapsed time is less than a first threshold, and a first operation to be performed when the elapsed time is equal to or greater than a first threshold. A second operation to be performed when the elapsed time is less than a second threshold value, and a third operation to be performed when the elapsed time exceeds the second threshold value. The utterance button switch is an operation that functions as an utterance button, and the first operation and the third operation are operations in which the utterance button switches function as different operation buttons.
[0017]
In the invention set forth in claim 11, in the operation setting storage processing, an operation is stored for each scene when a dialogue is performed, and the first threshold may be uttered in a target scene. It is a time shorter than the time required to utter the shortest response content of the response content, and the second threshold value is the most time responsive among the response content likely to be uttered in the target scene. It is characterized in that the time is longer than a time required for uttering a long response content.
[0018]
A twelfth aspect of the present invention is characterized in that the voice interaction program further causes a computer to perform an input process for arbitrarily setting the first and second thresholds.
[0019]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, a spoken dialogue system according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing the configuration of the embodiment. In this figure, reference numeral 1 denotes an utterance button attached to a steering wheel of an automobile. The utterance button is a button switch capable of discriminating between a push state and a release state, and the driver can operate the utterance button without releasing the hand from the steering wheel. Reference numeral 2 denotes an utterance button operation detection unit that detects two states of the utterance button 1 (push and release). Reference numeral 3 denotes a time monitoring unit that measures the elapsed time from when the utterance button 1 is pushed to when it is released, based on the output of the utterance button operation detection unit 2. Reference numeral 4 is for performing device control processing (such as turning on / off an air conditioner or playing / stopping an audio device) and information providing processing (such as executing a database search and reading out a search result) while interacting with the driver. An HMI (Human-Machine Interface) control unit that outputs a control signal to a device. Reference numeral 5 denotes a button operation setting storage unit in which an operation according to the operation state of the utterance button 1 is set in advance. Reference numeral 6 denotes a microphone for collecting sound. Reference numeral 7 denotes a voice recognition unit that recognizes voice collected by the microphone 6. Reference numeral 8 denotes a speaker that emits a sound produced by the system. Reference numeral 9 denotes an audio reproducing unit that reproduces audio data and causes the speaker 8 to generate sound. Reference numeral 10 denotes a voice synthesis unit that synthesizes voice data generated by the system. Reference numeral 11 denotes a dialogue control unit that controls the flow of a scenario when a dialogue is performed with the driver. Reference numeral 12 denotes a dialogue scenario storage unit in which a scenario for performing a dialogue with the driver is defined in advance. Reference numeral 13 denotes an input unit including operation switches such as a touch panel.
[0020]
Next, the operation of the voice interaction system shown in FIG. 1 will be described with reference to FIGS. In the following description, it is assumed that the HMI control unit 4 performs an information providing process of searching for a restaurant in a car and providing result information to a driver.
First, with reference to FIG. 2, an operation when the driver inputs a spoken voice will be described. First, the HMI control unit 4 reads the content of the operation setting stored in the button operation setting storage unit 5 when the voice dialogue system is activated (Step S0). This operation is executed only once at startup.
[0021]
Here, the contents stored in the button operation setting memory 5 will be described with reference to FIG. The button operation setting memory 5 stores the operation to be performed by the HMI controller 4 for each application (here, database search, air conditioner, etc.) when the utterance button 1 is “pushed” and when the utterance button 1 is “released”. Control, audio control). In the case of “release”, an operation that is subdivided for each elapsed time from “push” to “release” is further defined. In this example, when “push” is performed, an instruction to stop guidance reproduction and reproduce a chime sound is issued to the audio reproduction unit 9, and an instruction to start recording audio is issued to the audio recognition unit 7. It is defined that In the database search, the operation is defined every time the elapsed time until “release” is 0 to 200 msec, 201 to 10000 msec, 10001 msec or more. Details of the operation based on this table reference will be described later.
[0022]
This 200 msec is the first threshold value in the claims, and 10000 msec is the second threshold value. The first and second threshold values differ for each application, and the operation is defined in advance for each elapsed time in order to facilitate the operation. However, the first threshold value is a time shorter than a time required for uttering the shortest phrase (for example, “yes”) among the phrases that the speaker may utter in the target application. Set to. In addition, the second threshold value is set to a value that is longer than the time required to speak the longest phrase (eg, “Japanese restaurant search”) among the phrases that the speaker may possibly speak in the target application. Set a long time.
[0023]
Next, returning to FIG. 2, the driver selects a database search by operating a switch of the input unit 13 or by voice input operation. In response to the selection of the database search process, the dialog control unit 11 instructs the HMI control unit 4 to perform voice input. Then, the system issues a voice question "What data do you want to search?", And the HMI control unit 4 enters a voice input waiting state.
[0024]
Here, it is assumed that the driver pushes the utterance button 1 and utters “meal search” for, for example, six seconds (6000 msec), and then releases the utterance button 1. In response to this, the utterance button operation detection unit 2 detects that the utterance button 1 has been pushed, and sends an utterance button push detection notification to the time monitoring unit 3 (step S1). In response to this, the time monitoring unit 3 starts timer measurement (step S2) and sends an utterance button push detection notification to the HMI control unit 4 (step S3).
[0025]
Next, the HMI control unit 4 performs a push process according to the operation settings read in advance (step S4). In the process at the time of the push, first, in order to stop the reproduction of the guidance sound, the audio reproduction unit 9 is instructed to stop the guidance (step S5), whereby the guidance is stopped (step S6). Then, in order to notify the driver that the push operation has been detected by the system, the sound reproducing unit 9 is instructed to reproduce the chime (step S7), whereby the chime sound "pi" is reproduced. (Step S8). In addition, the voice recognition unit 7 issues an instruction to start voice recording to the voice recognition unit 7 (step S9), whereby the voice recognition unit 7 starts recording the voice collected by the microphone 6 and simultaneously performs the voice recording. Is performed (step S10).
[0026]
Next, when detecting the release of the utterance button 1, the utterance button operation detection unit 2 sends an utterance button release detection notification to the time monitoring unit 3 (step S11). Upon receiving this notification, the time monitoring unit 3 ends the timer measurement (step S12), obtains the elapsed time from the push to the release, and sends a speech button release detection notification including the elapsed time to the HMI control unit 4 (step S12). S13).
[0027]
Next, the HMI control unit 4 performs a process at the time of release based on the notified elapsed time (step S14). Here, since the elapsed time is 6000 msec, the operation of 201 to 10000 msec defined in the button operation setting storage unit 5 shown in FIG. 4 is performed. If the elapsed time is from 201 to 10000 msec, the HMI control unit 4 requests a recognition result from the speech recognition unit 7 already in the execution state (step S15), and obtains the recognition result (step S16). Then, the obtained speech recognition result (here, “meal search”) is notified to the dialogue control unit 11 (step S17).
[0028]
In response to this, the dialog control unit 11 refers to the dialog scenario storage unit 12 and reads the next operation scenario to be performed when “meal search” is specified (step S18). Then, based on the read scenario, the contents of the guidance to be uttered next are generated (step S19). In this example, the guidance content to be uttered next is “meal search”. Subsequently, the dialogue control unit 11 notifies the HMI control unit 4 of the guidance content "meal search" generated here (step S20).
[0029]
Next, the HMI control unit 4 sends a voice synthesis request including the notified guidance content to the voice synthesis unit 10 (Step S21). In response to this, the voice synthesis unit 10 generates voice data of the guidance of "meal search" by voice synthesis (step S22), and returns a voice synthesis result response including the voice data to the HMI control unit 4 (step S22). Step S23).
[0030]
Next, the HMI control unit 4 sends a sound reproduction request including the sound data to the sound reproduction unit 9 (Step S24). In response to this, the audio reproduction unit 9 reproduces the transmitted audio data (Step S25). As a result, a voice saying “meal search” is uttered from the speaker 8. Then, the audio reproduction unit 9 sends a guidance audio reproduction end notification to the HMI control unit 4 (Step S26).
[0031]
Subsequently, when the driver pushes the utterance button 1 in response to the question of "meal search", utters "yes" and releases the utterance button 1, processing is performed in the same manner as the above-described operation. , The dialog scenario will continue.
[0032]
Next, an operation when the driver operates the speech button 1 and responds instead of speaking will be described with reference to FIG. Here, it is assumed that the speaker 8 has uttered the question “meal search”. Here, it is assumed that the driver pushes the utterance button 1 and immediately releases the utterance button 1 (after 150 msec). In response to this, the utterance button operation detection unit 2 detects that the utterance button 1 has been pushed, and sends an utterance button push detection notification to the time monitoring unit 3 (step S1). In response to this, the time monitoring unit 3 starts timer measurement (step S2) and sends an utterance button push detection notification to the HMI control unit 4 (step S3).
[0033]
Next, the HMI control unit 4 performs a push process according to the operation settings read in advance (step S4). In the process at the time of the push, first, in order to stop the reproduction of the guidance sound, the audio reproduction unit 9 is instructed to stop the guidance (step S5), whereby the guidance is stopped (step S6). Then, in order to notify the driver that the push operation has been detected by the system, the sound reproducing unit 9 is instructed to reproduce the chime (step S7), whereby the chime sound "pi" is reproduced. (Step S8). In addition, the voice recognition unit 7 issues an instruction to start voice recording to the voice recognition unit 7 (step S9), whereby the voice recognition unit 7 starts recording the voice collected by the microphone 6 and simultaneously performs the voice recording. Is performed (step S10).
[0034]
Next, when detecting the release of the utterance button 1, the utterance button operation detection unit 2 sends an utterance button release detection notification to the time monitoring unit 3 (step S11). Upon receiving this notification, the time monitoring unit 3 ends the timer measurement (step S12), obtains the elapsed time from the push to the release, and sends a speech button release detection notification including the elapsed time to the HMI control unit 4 (step S12). S13). The operations in steps S1 to S13 shown in FIG. 3 are the same as the operations in steps S1 to S13 shown in FIG.
[0035]
Next, the HMI control unit 4 performs a process at the time of release based on the notified elapsed time (step S31). Here, since the elapsed time is 150 msec, the operation of 0 to 200 msec defined in the button operation setting memory 5 shown in FIG. 4 is performed. If the elapsed time is 0 to 200 msec, the HMI control unit 4 issues an instruction to the voice recognition unit 7 to stop the voice recognition process that has already been executed (step S32), and identifies the input as an operation button input. Is instructed to the sound reproducing unit 9 to notify the driver (step S33), whereby a chime sound “Po” is reproduced (step S34). Then, the control unit 11 notifies the dialog control unit 11 that the user input is an operation button input and the value is a short-time push (0 to 200 msec) (step S35). In response to this, the dialog control unit 11 refers to the dialog scenario storage unit 12 (step S36), and recognizes this input result (step S37). With this recognition, the input result notified in step S35 is regarded as having answered "Yes" by voice in response to the question of "meal search", and the subsequent processing is continued. Operations after step S37 are the same as steps S18 to S26 shown in FIG.
[0036]
When the elapsed time notified in step S13 shown in FIG. 3 is 10001 msec or more, the HMI control unit 4 reproduces a chime “popo” indicating that the chime is identified as an operation button input, and the HMI control unit 4 is already in the execution state. The voice recognition principle is canceled, and the dialog control unit 11 is notified that the user's input is an operation button input and the value is a long-time push.
[0037]
The HMI control unit 4 issues an instruction to the voice recognition unit 7 to stop the voice recognition process that has already been performed (step S32), and also notifies the driver that the voice recognition process has been identified as an operation button input. The reproduction unit 9 is instructed to reproduce the chime (step S33), and the chime sound “popo” is reproduced (step S34). Then, the control unit 11 notifies the dialog control unit 11 that the user input is an operation button input and the value has been a long-time push (10001 msec or more) (step S35). The long push is used when the currently running application (here, database search) is to be terminated halfway, or when it is desired to clear the processing contents up to the present and return to the initial state.
[0038]
As described above, the time monitoring unit 3 is provided, and the elapsed time from when the driver pushes the utterance button 1 until release is measured. When the elapsed time is short (200 msec or less) and long (10001 msec or more). Since it is determined that the utterance button 1 has been used as an operation button and the operation defined in the button operation operation setting storage unit 5 is performed, a press talk function for voice input can be performed with one utterance button. An operation button function for realizing a shortcut of an interactive operation or the like can be realized at the same time, and the user can select an input that is convenient for the user. As a result, in a situation where the answer to the inquiry to the driver is “yes / no”, the short-time push is regarded as a voice input of “yes”, and the answer to the inquiry to the driver is plural. In a scene where selection is from a candidate, it is possible to easily perform state transition control by regarding short-time push as "again" voice input.
[0039]
In addition, voice recognition may cause erroneous recognition due to driving noise, etc., whereas shortcuts by button operation are reliable, and dialogue can be performed in a short time by replacing frequently used voice input commands with shortcuts. You can proceed. Therefore, the user operation can be simplified and the load can be reduced, and the manufacturing cost can be reduced by reducing the number of operation buttons.
[0040]
In the above description, in step S0 shown in FIG. 2, all the contents of the button operation setting storage unit 5 shown in FIG. 4 are read at the time of starting up the system. Only the contents of the table may be read. For example, when a database search application is started, only the operation setting for database search is read from the button operation setting storage unit 5. Thus, the time for reading the contents of the setting operation can be reduced. In the example shown in FIG. 4, one table is prepared for one application. However, one table is prepared for each scene of one application, and operation is performed for each scene. The meaning may be changed. Also, in the above description, an example of answering only by voice is shown, but a dialogue means other than voice is used together, such as a touch panel display device in which display contents and selection candidate buttons change according to the progress of the dialogue. Is also good.
[0041]
Further, the contents of the button operation setting memory 5 shown in FIG. 4 may be rewritable based on the contents inputted by the driver through the input unit 3 by the driver. Here, rewriting includes new addition, deletion, and change. Thereby, the first threshold value of the elapsed time considered as a short-time push is set to 300 msec instead of 200 msec, and the second threshold value of the elapsed time considered to be a long-time push is set to 8000 msec instead of 10001 msec. Therefore, it is possible for the driver to easily operate and customize. However, as described above, the first threshold is set to a time shorter than the time required to speak the shortest phrase spoken by the speaker, and the second threshold is set to the time when the speaker speaks. The time must be set to be longer than the time required to speak the longest phrase.
[0042]
In the above description, the navigation system has been described as an example. However, the above-described method can be similarly applied to any device that enables a voice conversation with the system using a press talk input.
[0043]
Also, a program for realizing the functions of the respective processing units in FIG. 1 is recorded on a computer-readable recording medium, and the program recorded on this recording medium is read into a computer system and executed, thereby executing a voice interactive processing. May be performed. Here, the “computer system” includes an OS and hardware such as peripheral devices. The “computer-readable recording medium” refers to a portable medium such as a flexible disk, a magneto-optical disk, a ROM, and a CD-ROM, and a storage device such as a hard disk built in a computer system. Further, the “computer-readable recording medium” refers to a volatile memory (RAM) inside a computer system that becomes a server or a client when a program is transmitted via a network such as the Internet or a communication line such as a telephone line. In addition, those that hold programs for a certain period of time are also included.
[0044]
Further, the above program may be transmitted from a computer system storing the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in the transmission medium. Here, the "transmission medium" for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line. Further, the program may be for realizing a part of the functions described above. Furthermore, what can implement | achieve the function mentioned above in combination with the program already recorded on the computer system, and what is called a difference file (difference program) may be sufficient.
[0045]
【The invention's effect】
As described above, according to the present invention, the time monitoring means is provided, and the elapsed time from when the user pushes the utterance button to when the utterance button is released is measured. Is determined to be used as an operation button, and the operation defined in the button operation operation setting storage unit is performed.Therefore, with a single utterance button, a press-talk function for voice input and a shortcut for an interactive operation are provided. The operation button function for realizing can be realized at the same time, and the user can select an input that is convenient for him. In a situation where the answer to the inquiry to the user is “yes / no”, the short push is regarded as a voice input of “yes”, and the answer to the inquiry to the driver is determined from a plurality of candidates. In such a case, it is possible to obtain an effect that it is possible to easily perform the state transition control by regarding the short-time push as the "again" voice input. Further, since the input means is provided so that the contents stored in the operation setting storage means can be changed, it is possible to customize the operation setting to an easy-to-use form for the user.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of an embodiment of the present invention.
FIG. 2 is a sequence diagram showing an operation at the time of voice input in the voice dialogue system shown in FIG. 1;
FIG. 3 is a sequence diagram showing an operation when a button is operated in the voice interaction system shown in FIG. 1;
FIG. 4 is an explanatory diagram showing a table structure of a button operation setting memory 5 shown in FIG. 1;
[Explanation of symbols]
1 ... speak button
2 ... utterance button operation detection unit
3. Time monitoring unit
4 ... HMI control unit
5 ... button operation setting memory
6 ... microphone
7 ... Speech recognition unit
8 ... speaker
9 Voice playback unit
10 Voice synthesis unit
11 Dialogue control unit
12: Dialogue scenario storage unit

Claims (12)

A speech dialogue system including a speech button switch for pushing only while the speaker is speaking, and controlling the device while the speaker interacts with the system by voice,
Time monitoring means for measuring an elapsed time from when the utterance button switch is pushed to when it is released,
An operation setting storage unit in which an operation to be performed next for each elapsed time is stored in advance;
A voice dialogue system comprising: an operation corresponding to an elapsed time notified by the time monitoring unit, read from the operation setting storage unit, and a control unit that executes the read operation. The operation setting storage means,
A first operation to be performed when the elapsed time is less than a first threshold, and a second operation to be performed when the elapsed time is equal to or more than a first threshold and less than a second threshold. A third operation to be performed when the elapsed time exceeds a second threshold value is stored in advance; the second operation is an operation in which the utterance button switch functions as an utterance button; 2. The voice interaction system according to claim 1, wherein the third operation and the third operation are operations in which the speech button switches function as different operation buttons. The operation setting storage means,
An operation is stored for each scene when a dialogue is performed, and the first threshold value is necessary for uttering the shortest response content among the response contents that may be uttered in the target scene. And the second threshold is longer than the time required to utter the longest response content among the response contents that may be uttered in the target scene. 3. The voice interaction system according to claim 2, wherein: The voice interaction system,
4. The voice interaction system according to claim 2, further comprising an input unit for arbitrarily setting the first and second thresholds. A speech dialogue method in a speech dialogue system comprising a speech button switch for pushing only while a speaker is speaking, and controlling a device while the speaker interacts with the system by voice.
A time monitoring step of measuring the elapsed time from when the utterance button switch is pushed to when it is released,
An operation setting storage step of storing in advance an operation to be performed next for each elapsed time;
A control step of reading and executing an operation corresponding to the elapsed time notified by the time monitoring step from the operations stored in the operation setting storage step. The operation setting storing step includes:
A first operation to be performed when the elapsed time is less than a first threshold, and a second operation to be performed when the elapsed time is equal to or more than a first threshold and less than a second threshold. Storing in advance a third operation to be performed when the elapsed time exceeds a second threshold value, wherein the second operation is an operation in which the utterance button switch functions as an utterance button; The voice interaction method according to claim 5, wherein the third operation and the third operation are operations in which the utterance button switches function as different operation buttons. The operation setting storing step includes:
An operation is stored for each scene when a dialogue is performed, and the first threshold value is necessary for uttering the shortest response content among the response contents that may be uttered in the target scene. And the second threshold is longer than the time required to utter the longest response content among the response contents that may be uttered in the target scene. The voice interaction method according to claim 6, wherein: The voice interaction method includes:
The method according to claim 6, further comprising an input step of arbitrarily setting the first and second thresholds. A speech dialogue program comprising a speech button switch for pushing only while the speaker is speaking, and operating in a speech dialogue system for controlling a device while the speaker interacts with the system by voice,
A time monitoring process for measuring an elapsed time from when the utterance button switch is pushed to when it is released,
An operation setting storage process for storing in advance an operation to be performed next for each elapsed time;
A voice dialogue program for causing a computer to perform a control process of reading and executing an operation corresponding to an elapsed time notified by the time monitoring process from the operations stored in the operation setting storage process. The operation setting storage process includes:
A first operation to be performed when the elapsed time is less than a first threshold, and a second operation to be performed when the elapsed time is equal to or more than a first threshold and less than a second threshold. Storing in advance a third operation to be performed when the elapsed time exceeds a second threshold value, wherein the second operation is an operation in which the utterance button switch functions as an utterance button; 10. The voice dialogue program according to claim 9, wherein the third operation and the third operation are operations in which the speech button switches function as different operation buttons. The operation setting storage process includes:
An operation is stored for each scene when a dialogue is performed, and the first threshold value is necessary for uttering the shortest response content among the response contents that may be uttered in the target scene. And the second threshold value is set to a time longer than the time required to utter the longest response content among the response contents that may be uttered in the target scene. The speech dialogue program according to claim 10, wherein the program is set. The voice dialogue program,
The speech dialogue program according to claim 10 or 11, further causing a computer to perform an input process for arbitrarily setting the first and second thresholds.

Images