JP3846161B2 - Information provision system by voice and its malfunction cause notification method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a voice information providing system that receives a voice command from a user and can provide information corresponding to the voice command to the user by voice, and a malfunction cause notification method thereof.
[0002]
[Prior art]
In recent years, information processing systems using speech have been proposed in various fields due to advances in speech-related technologies such as speech recognition technology and speech synthesis technology. For example, commands that can be used to input commands to the system by voice and that can respond from the system by voice have already been developed. Such a voice information processing system is particularly convenient for users with visual disabilities and users with limited limbs, and is expected to spread in various fields.
[0003]
As one example, a life support system for visually handicapped persons by voice response has been proposed (Institute of Electronics, Information and Communication Engineers, Type 2 Study Group, pages 33-39, "Life for visually handicapped persons by voice response" Prototype support system ”Shizuoka Prefectural University Faculty of Business and Information Sciences Kenichi Matsuura Hiroaki Yuse).
[0004]
This system is schematically configured as shown in FIG. 1, and the user U wears the headphones 32 and the microphone 11 and uses the communication means such as wireless communication to transmit audio information to and from the system side. It is possible to exchange and acquire various information by voice. The dialogue example is shown below. The user is represented by U, and the system is represented by Sy. In addition, the system Sy referred to here is a generic term for the user terminal T, the host computer H, and the like in FIG.
[0005]
U: Clock.
Sy: You can use the clock / calendar function.
U: What time is it now?
Sy: The current time is 4:30 pm.
U: What day is next Thursday?
Sy: Next Thursday is May 10th.
U: Calculate with a calculator
Sy: You can use the calculator function.
U: 980 plus 198
Sy: 980 plus 198 is 1178.
U: Dictionaries
Sy: You can use the dictionary function.
U: “Active”
Sy: Is the word you're looking for "Sekkikuki"?
U: Yes.
[0006]
In this way, the user U issues a command for acquiring information by voice, the system Sy recognizes the voice command, and returns an answer based on the recognition result to the user U by voice. Furthermore, it is also possible to obtain information by inquiring about life information such as weather forecasts, news, timetables such as railways, and television programs. In this case, a specific web page is designated in advance, and in response to a request from the user U, the host computer H connects to the server S shown in FIG. 1 and the server S acquires information from the web page. This is notified to the user U.
[0007]
[Problems to be solved by the invention]
The above-described information providing system by voice is expected to be useful not only for visually impaired users but also for healthy people, but there are still many problems to be solved.
[0008]
For example, there are many cases where there is no response from the system Sy side to the voice command input by the user U, or there is a totally wrong response. This is not only a problem in the speech recognition performance and application on the system Sy side, but there may be a problem in the way the user speaks, the microphone 11 and the like, and there may be a problem in the communication environment. In this way, malfunctions (hereinafter referred to as response abnormalities) such as no response from the system Sy side or an incorrect response are caused by various causes.
[0009]
If the user knows the cause of the response abnormality, it can cope with it. For example, if the cause is known such as the volume of the spoken voice is too low, the voice is spoken too quickly, or the power switch of the microphone is not turned on, it is possible to cope with it.
[0010]
However, this type of conventional system usually does not have a means to inform the user of the cause, and it is very difficult for the user to identify the cause of the abnormal response, Eventually, it is necessary to perform a trial and error cause diagnosis such as repeating the same utterance or examining the state of the system. If you are a healthy person, you may be able to identify the cause by investigating the time using various methods, but it is difficult for users with visual impairments to investigate the cause. It is necessary to have a specialist in this type of system investigate the cause.
[0011]
Therefore, the present invention makes it possible to self-diagnose what is the cause when the system malfunctions in response to a voice command and to inform the user of the self-diagnosis result in an easy-to-understand manner. By doing this, it is an object to provide a system that is easy to use not only for healthy people but also for users with visual impairments.
[0012]
[Means for Solving the Problems]
In order to achieve the above-described object, the voice information providing system according to the present invention receives a voice command from the terminal owned by the user and receives the voice command on the host computer side, and the host computer receives the voice command. The voice command is recognized and a response to the recognition result is transmitted to the terminal. If the response from the host computer to the voice command is abnormal, the cause of the abnormal response is given to the user. A voice information providing system having a means for informing the user, and when there is a response abnormality in the response from the host computer to the voice command, a means for notifying the user of the cause of the response abnormality, When a user detects an abnormal response on the host computer side, the user A self-diagnosis start signal detecting means for obtaining a self-diagnosis start signal by detecting an operation performed by the self-diagnosis start signal detecting means, a self-diagnosis means for starting a self-diagnosis upon receiving a self-diagnosis start instruction signal from the self-diagnosis start signal detecting means, It has a configuration having notifying means for notifying the user of the self-diagnosis result obtained by the self-diagnosis means.
[0013]
In such a voice information providing system, the response from the host computer and the notification of the self-diagnosis result to the user are both performed by voice in response to the voice command.
[0014]
The operation performed by the user necessary for obtaining the self-diagnosis start signal is an operation in which the user confirms the operation state of the microphone with respect to the microphone existing in the terminal.
[0015]
Here, the user confirming the operation state of the microphone is at least one of an operation in which the user blows the microphone, an operation in which the user inputs a predetermined sound into the microphone, and an operation in which the user taps the microphone. It is conceivable to use an operation.
[0016]
The self-diagnosis start signal detection means, the self-diagnosis means, and the notification means are mainly provided on the terminal side.
[0017]
In addition, when there is a response abnormality in the response from the host computer side to the voice command, the voice command input by the user can be fed back to the user.
[0018]
Further, the malfunction cause notification method of the information providing system by voice according to the present invention is such that the user inputs a voice command from a terminal possessed by the user, the voice command is received by the host computer, and the host computer receives the voice command. A means for recognizing a voice command and transmitting a response to the recognition result to the terminal, and when there is a response abnormality in the response from the host computer to the voice command, the cause of the response abnormality to the user A malfunction cause notifying method of a voice information providing system having means for informing the user, and when there is a response abnormality in a response from the host computer to the voice command, the cause of the response abnormality is notified to the user Means for the user to detect an abnormal response on the host computer side. The self-diagnosis start motion signal is obtained by detecting the operation performed by the user as a self-diagnosis start motion signal, and the self-diagnosis function is activated thereby to investigate the cause of the response abnormality. Diagnosis is performed and the self-diagnosis result is notified to the user.
[0019]
In such a malfunction cause notification method of the information providing system by voice, both the response from the host computer and the notification of the self-diagnosis result to the voice command are performed by voice.
[0020]
The operation performed by the user necessary for obtaining the self-diagnosis start signal is an operation in which the user confirms the operation state of the microphone with respect to the microphone existing in the terminal.
[0021]
Here, the user confirming the operation state of the microphone is at least one of an operation in which the user blows the microphone, an operation in which the user inputs a predetermined sound into the microphone, and an operation in which the user taps the microphone. It is conceivable to use an operation.
[0022]
Then, the function of obtaining the self-diagnosis start signal, the function of performing the self-diagnosis, and the function of notifying the user of the self-diagnosis result are mainly provided on the terminal side.
[0023]
In addition, even in this voice information providing system malfunction cause notification method, if there is a response abnormality in the response from the host computer to the voice command, the voice command input by the user can be fed back to the user. .
[0024]
As described above, in the present invention, when a host computer (hereinafter simply referred to as a host) causes an abnormal response to a voice command input by the user, an operation performed by the user when the user senses that there is an abnormal response. Thus, the self-diagnosis function is activated on the terminal side, the cause of the abnormality is examined by the self-diagnosis function, and the result is presented to the user.
[0025]
In other words, the response from the host computer to the voice command issued by the user is clearly an abnormal response (this response abnormality is, as described above, no response to the voice command issued by the user, When the user determines that the operation is erroneous (such as an incorrect response), the user normally performs some action naturally. In the system, by detecting the operation of the user, self-diagnosis is performed as to where the response is abnormal and the self-diagnosis result is notified to the user.
[0026]
In this way, when the user takes some action for the response abnormality, the cause of the response abnormality can be self-diagnosed and investigated, and the result of the self-diagnosis can be notified to the user. There is no need to investigate each cause. In addition, at this time, if the self-diagnosis result is notified by voice, it becomes a convenient system for a visually impaired user, and even a healthy person can obtain the diagnosis result from the ear while on the spot. Therefore, it is convenient when using this information providing system while doing other work.
[0027]
The operation performed by the user when he / she knows that there is a response abnormality is an operation in which the user confirms the operation state of the microphone with respect to the microphone existing on the user side. Is the action of blowing a breath like "Foot" on the microphone, the action of inputting a sound like "Ah", the action of tapping the microphone, etc. These actions try to speak toward the microphone This is a natural operation when the user feels something abnormal.
[0028]
As described above, since the natural operation performed by the user for the response abnormality is used as a signal for starting the self-diagnosis function, the user can automatically activate the self-diagnosis function without performing a special operation.
[0029]
It should be noted that a function for obtaining a signal for starting the self-diagnosis function, a function for performing self-diagnosis, and a function for notifying the user of the self-diagnosis result are mainly provided on the terminal side. As a result, when the present invention is applied to an existing system such as the “life support system for visually impaired persons by voice response” described as an example of the above-described conventional information providing system, the host side is significantly changed. Since the present invention can be realized without any problem, this kind of existing information providing system can be effectively used.
[0030]
In addition, when self-diagnosis occurs when a response abnormality occurs, the voice command entered by the user can be fed back to the user as it is, and the voice command entered by the user is fed back as it is. You can know the status of the command. For example, if the feedback voice command shows that the voice is too loud and the sound is broken, it means that the user was unable to properly recognize the voice because his / her voice was too loud. Understand and know that you should speak carefully with that point. By repeating this several times, the user can learn how to speak properly to be recognized.
[0031]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described below.
[0032]
FIG. 1 is a diagram showing a schematic configuration of an information providing system by voice. Broadly speaking, there are a terminal T on a user side and a host computer (simply referred to as a host as described above) H. Depending on the content of the voice command, information may be acquired from, for example, a server S connected to the network N as an external information providing unit.
[0033]
As shown in FIG. 2, the user-side terminal T includes a voice input processing unit 1, a self-diagnosis processing unit 2, a voice output processing unit 3, a wireless communication unit 4, and the like. Each of these components will be described in detail later.
[0034]
The host H includes a wireless communication unit 61 that enables wireless communication with the terminal T, a voice recognition unit 62 that recognizes a voice transmitted from the terminal T via the wireless communication unit 61, and at least the function of the present invention. An application unit 63 composed of several applications necessary for executing, a voice synthesis unit 64 for generating a voice signal for responding by voice to the terminal T, and a network communication unit for acquiring information from the server S 65 and the like.
[0035]
The server S includes a network communication unit 71, a database 72 in which various data are stored, and the like.
[0036]
FIG. 3 is a diagram for further explaining each component of the terminal 1. The voice input processing unit 1 includes a microphone 11, a voice input unit 12, a voice temporary storage unit 13, a voice section detection unit 14, and the like. The self-diagnosis processing unit 2 includes a self-diagnosis start signal detection unit 21, a self-diagnosis unit 22, an error message storage unit 23, and the like, and the audio output processing unit 3 includes an audio output unit 31, headphones 32, and the like. . The wireless communication unit 4 includes a signal transmission unit 41 and a signal reception unit 42. The user U wears the microphone 11 and the headphones 32 as shown in FIG.
[0037]
Note that although sound other than sound may be input to the microphone 11, the sound input processing unit 1 can of course process sound other than sound.
[0038]
A sound signal output from the microphone 11 (here, a sound signal including a signal other than the sound signal is also referred to as a sound signal) is a general sound signal such as an A / D conversion process or an amplification process in the sound input unit 12. After the processing, it is temporarily stored in the audio temporary storage unit 13 sequentially for every certain interval.
[0039]
Then, the sound signal from the microphone 11 (referred to as an input sound signal) stored in the sound temporary storage unit 13 is read by the self-diagnosis start signal detection unit 21 to determine whether or not it is a self-diagnosis start signal. Made. As described above, the self-diagnosis start signal detection unit 21 detects a signal obtained by an operation performed by the user when the user senses a response abnormality, for example, an operation in which the user confirms the operation state of the microphone 11. Specifically, at least one of the operation in which the user blows the microphone 11 like “Foot”, the operation of inputting the sound “Ah”, and the operation of tapping the microphone Detects a signal obtained by being made by the user.
[0040]
The self-diagnosis start signal detection process performed by the self-diagnosis start signal detector 21 will be described with reference to the flowchart of FIG.
[0041]
In FIG. 4, first, the input voice signal temporarily stored in the voice temporary storage unit 13 is subjected to signal analysis to obtain a signal pattern of the input voice signal (step s1). Then, several types of self-diagnosis start signal patterns prepared in advance as self-diagnosis start signals are stored in the self-diagnosis start signal pattern storage unit 211 (the self-diagnosis start signal pattern storage unit 211 is stored in the self-diagnosis start signal detection unit 21). And pattern matching is performed with the signal pattern of the input audio signal obtained by signal analysis (step s2).
[0042]
In this case, the self-diagnosis start signal pattern registered in the self-diagnosis start signal pattern storage unit 211 is a signal obtained when the user performs an operation of blowing a breath such as “Foot” on the microphone 11. A pattern (referred to as a first diagnosis start signal pattern P1), a signal pattern obtained when the user performs an operation of inputting a sound such as “ah” toward the microphone 11 (referred to as a second diagnosis start pattern P2) And a signal pattern (referred to as a third diagnosis start pattern P3) obtained when the user performs an operation of tapping the microphone with a fingertip.
[0043]
Such various diagnosis start signal patterns P1, P2, P3 and input voice signal patterns are individually pattern-matched, and the input voice signal pattern corresponds to one of the diagnosis start signal patterns P1, P2, P3, That is, it is determined whether or not the input voice signal is a self-diagnosis start signal (step s3). In this determination, if it is determined that the input sound signal is a self-diagnosis start signal, the self-diagnosis process is started (step s4), and if it is determined that the input sound signal is not a self-diagnosis start signal. Then, the current self-diagnosis start detection process is terminated.
[0044]
Then, when it is determined that the input voice signal is a self-diagnosis start signal, the self-diagnosis process is performed by the self-diagnosis unit 22, and first, the terminal T performs self-diagnosis, and then the host H performs self-diagnosis, Further, self-diagnosis is sequentially performed on the network N and the server S. At this time, when performing self-diagnosis after the host H, a self-diagnosis instruction signal from the self-diagnosis unit 22 is sent to the host H side by the signal transmission unit 41 of the wireless communication unit 4.
[0045]
On the host H side, self-diagnosis is performed by starting a self-diagnosis application present in the application unit 63 by a self-diagnosis instruction signal sent from the terminal T side. A specific example of this self-diagnosis will be described later.
[0046]
If the cause of the response abnormality can be identified as a result of the self-diagnosis, the self-diagnosis unit 22 reads out an error message corresponding to the cause of the response abnormality from the error message storage unit 23 and sends it to the voice output unit 31. The audio output unit 31 gives the read error message to the headphones 32 as an audio signal. The error message storage unit 23 is prepared with error messages corresponding to various causes that cause a response abnormality.
[0047]
For example, an error message such as “There is a situation where radio waves are difficult to reach” is prepared as an error message corresponding to the case where the communication environment is bad, and the input voice is too low. An error message such as “Your voice is too low” is prepared as an error message corresponding to the case where the recognition is not performed normally. As described above, various error messages are prepared corresponding to various causes.
[0048]
On the other hand, if the self-diagnosis start signal detection unit 21 determines whether or not the input voice signal is a self-diagnosis start signal, if it is determined that it is not a self-diagnosis start signal, it is a voice command. Regardless, the process is passed to the speech segment detection unit 14. As a result, the voice segment detection unit 14 detects a voice segment from the input voice signal stored in the voice temporary storage unit 13 and sends a voice signal corresponding to the voice segment to the wireless communication unit 4. The wireless communication unit 4 transmits the audio signal from the signal transmission unit 41 to the host H side.
[0049]
Then, on the host H side, the voice signal sent from the terminal T side is received by the wireless communication unit 61, the voice recognition unit 62 performs voice recognition processing, and the application unit 63 responds based on the recognition result. Select and launch an application to perform processing corresponding to the input voice signal (voice command).
[0050]
If the voice command at this time is content that can be processed by the host H (for example, the voice command can be handled by using a clock or calendar function, the content that can be handled by using a dictionary function, etc.) A response corresponding to the command is generated by the speech synthesizer 64 and transmitted to the terminal T side by the wireless communication unit 61.
[0051]
Further, if the voice command is content that needs to acquire information from the server S (the voice command is content that needs to acquire information from a specific server such as weather forecast or news of the day, for example), the network communication unit 65 To obtain a desired information from the server S, generate a response content corresponding to the voice command from the user by the voice synthesizer, and transmit it to the terminal T side by the wireless communication unit 61 . On the terminal T side, the audio signal transmitted from the host H side is received by the signal receiving unit 42 of the wireless communication unit 4, processed by the audio output unit 31, and then output as sound from the headphones 32.
[0052]
Next, a specific operation of the voice information providing system of the present invention will be described. As described above, this type of voice information providing system is such that the user provides information desired by the user by voice when the user makes an inquiry by voice. If there is a problem somewhere in the information providing system by voice and the system malfunctions, that is, if the response from the host H becomes abnormal (this response abnormality is Despite giving a voice command, this means a malfunction such as no response from Host H or incorrect response). It is easy to convey by voice.
[0053]
Here, there are various causes for the response abnormality. That is, the cause of the response abnormality may be on the terminal T side, may be on the host H side, and may be on the network N or server S side.
[0054]
Causes on the terminal T side include, for example, a problem with a power supply unit (battery, switch, etc.), a problem with communication, a problem with an input / output system (microphone 11 or headphone 31), a problem such as how to input voice commands performed by the user And so on. Examples of problems in the way voice commands are input by the user include, for example, when the voice of the user's utterance is too large or too small, or when the utterance is too far or too close to the microphone 11 Or a word other than a recognizable word is input.
[0055]
Further, the cause on the host H side is a problem with the computer itself, a problem with the recognition performance of the voice recognition unit 62, a problem with the application 63, a problem with communication, and when acquiring information from the network N, There may be a case where there is a problem in connection, a problem on the line, or a problem in the network N or the server S itself.
[0056]
As described above, there are various causes for the host H to cause an abnormal response. Even if the response is abnormal due to the various causes described above, the user often does not know what caused the response abnormality.
[0057]
For example, when the user asks the host H "What time is it now?", Even if there is no response from the host H side or a response that is completely misplaced, the user may not be able to determine the cause. Many. In such a case, the user generally performs an operation of confirming the operation state of the microphone 11 naturally. Specifically, it blows on the microphone “Foot”, inputs a sound such as “Ah”, or taps the microphone 11 lightly.
[0058]
Thus, when at least one operation is performed among several operations for confirming the operation state of the microphone 11 that the user naturally performs when the user senses that there is a problem, the terminal T is provided. The detected self-diagnosis start signal detection unit 21 detects this and activates the self-diagnosis unit 22. The self-diagnosis start signal detection process performed by the self-diagnosis start signal detector 21 has already been described with reference to FIG.
[0059]
When the self-diagnosis start signal detection unit 21 determines that the input voice is a self-diagnosis start signal, the self-diagnosis unit 22 first self-diagnose the terminal T, then self-diagnose the host H, and then Then, the self-diagnosis is performed in the order of the terminal T, the host H, the network N, and the server S, as the network N and the server S are self-diagnosed.
[0060]
For example, when there is no particular problem on the terminal T side, the self-diagnosis unit 22 transmits a self-diagnosis instruction signal to the host H side by the wireless communication unit 4, and when the host H receives this self-diagnosis instruction signal, the host H On the H side, a self-diagnosis application in the application unit 63 is activated to perform self-diagnosis.
[0061]
Here, as a result of performing a self-diagnosis on the host H side, when it is found that when the speech recognition unit 62 performs speech recognition processing, the user's input speech is too small to perform proper speech recognition processing. The diagnosis result such as “Your voice is too low” is notified to the user by voice. In addition, if it is necessary to obtain information from the server S, if a response abnormality occurs, the network N and the server S are also diagnosed. For example, there is a problem with the connection to the network N. If known, the diagnosis result such as “There is a problem with the network connection” is notified to the user by voice.
[0062]
Thus, when it is determined that the input voice is a self-diagnosis start signal, the self-diagnosis unit 22 is activated and performs self-diagnosis in the order of the terminal T, the host H, the network N, and the server S, As a result, if a cause of response abnormality is found, the cause is specifically notified to the user by voice.
[0063]
As a result, even if the system side causes an abnormal response to the voice command issued by the user, the user can specifically know where and what problem caused the abnormality. In this way, if the cause of the response abnormality is known, it is possible to correct the cause of the abnormality and input the voice command again, eliminating waste of repeatedly inputting the voice command without knowing the cause. be able to.
[0064]
The present invention is not limited to the embodiment described above, and various modifications can be made without departing from the gist of the present invention.
[0065]
For example, when a response abnormality occurs with respect to a user's voice command and the system performs a self-diagnosis, the voice command input by the user may be fed back to the user as it is. As an example, the user tries to inquire about the current time, gives a voice command “What time is it now” to the host side, does not respond to the voice command, and the user blows on the microphone 11 with a “huff”. Suppose you have performed the correct operation. When the user performs such an operation of confirming the operation state of the microphone 11, when the self-diagnosis signal detection unit 21 detects it, the self-diagnosis unit 22 enters a self-diagnosis operation. The voice command portion (in this case, “What time is it now”) stored in the voice section is detected by the voice section detection unit 14, and the voice section is processed by the voice output unit 31 and output from the headphones 32.
[0066]
As a result, the voice command (“what time is it now”) spoken by the user is output as it is from the headphones 32. The user can know the state of the voice command inputted by listening to the user. For example, if it is "What time is it?" That you have spoken from the headphones 32, but the voice is too loud and the sound seems to be cracking, then the user will be able to speak properly because the voice you spoke is too loud. Know that you couldn't recognize, and know that you should speak carefully.
[0067]
In the above-described embodiment, the self-diagnosis start signal detection unit 21 tests the operation state of the microphone 11 by blowing the microphone 11, inputting a sound such as “Ah”, and tapping the microphone. In this case, the self-diagnosis start signal is obtained by detecting the operation generally performed by the user at the time of the operation. The operation generally performed by the user when testing the operation state of the microphone 11 is the above-described three examples. However, the present invention is not limited to these, and various other methods are conceivable.
[0068]
Further, the self-diagnosis start signal is not limited to being obtained by detecting an operation generally performed by the user when testing the operation state of the microphone 11 described above. For example, a self-diagnosis start button is prepared, When a response abnormality occurs in the system, a self-diagnosis start signal may be obtained by pressing the self-diagnosis start button.
[0069]
Also, some keyword may be prepared, and if the system causes a response abnormality, the user may speak the keyword. In this case, the self-diagnosis start signal detection unit 21 may detect that the keyword has been issued, and the self-diagnosis unit 22 may be activated accordingly.
[0070]
Furthermore, the error message presented to the user can be displayed not only by voice but also by text display on the terminal T on the user side, and voice and text display may be used together.
[0071]
In addition, a message that responds to the user from the system side, a message that responds to the user from the terminal T (in the case of the above-described embodiment, an error message, etc.) and a message that responds to the user from the host H (obtained for voice commands) Voice quality can be made different from each other). For example, by making the voice messages different from each other such that the error message is a male voice and the obtained information is a female voice, the user simply listens to the response and the response is from the terminal T. It is immediately known whether it is from H, and it is particularly convenient for the visually impaired.
[0072]
The processing procedure of the present invention described above can be recorded on a recording medium such as a floppy disk, an optical disk, or a hard disk. The present invention includes the recording medium. Further, the processing procedure may be obtained via a network.
[0073]
【The invention's effect】
As described above, according to the present invention, when the host computer side causes a response abnormality to a voice command input by the user, an operation performed by the user is detected when the user senses that there is a response abnormality. Thereby, a self-diagnosis function is activated on the terminal side, the cause of the abnormality is examined by the self-diagnosis function, and the result is presented to the user. In this way, when the user takes some action for the response abnormality, the cause of the response abnormality can be self-diagnosed and investigated, and the result of the self-diagnosis can be notified to the user. There is no need to investigate each cause. In addition, at this time, if the self-diagnosis result is notified by voice, it becomes a convenient system for a visually impaired user, and even a healthy person can obtain the diagnosis result from the ear while on the spot. Therefore, when using this information provision system while doing other work, it is possible to know what went wrong and the response abnormality occurred without resting the hands and eyes working.
[0074]
The operation performed by the user when he / she knows that there is a response abnormality is an operation in which the user confirms the operation state of the microphone with respect to the microphone existing on the user side. Is the action of blowing a breath like "Foot" on the microphone, the action of inputting a sound like "Ah", the action of tapping the microphone, etc. These actions try to speak toward the microphone This is a natural operation when the user feels something abnormal.
[0075]
As described above, since the natural operation performed by the user for the response abnormality is used as a signal for starting the self-diagnosis function, the user can automatically activate the self-diagnosis function without performing a special operation.
[0076]
Further, according to the present invention, when a response abnormality occurs and self-diagnosis is performed, the voice command input by the user can be fed back to the user as it is, so that the user knows the state of the voice command input by the user. By repeating this several times, the user can learn how to speak to properly recognize the voice.
[0077]
In this way, the present invention receives a voice command from a user and can provide information corresponding to the voice command mainly by voice. It is possible to self-diagnose the cause and inform the user of the self-diagnostic result in an easy-to-understand manner. In particular, the self-diagnosis result is presented to the user by voice. By applying the present invention to a life support system for persons with disabilities, the system can be made even easier for visually handicapped persons. In addition, not only visually impaired people but also healthy people are easy to use for users who are unfamiliar with the handling of this type of system or who have many opportunities to use such information provision systems while doing other work. It will be good.
[Brief description of the drawings]
FIG. 1 is a diagram showing an overall system configuration for explaining an embodiment of a voice information providing system according to the present invention.
FIG. 2 is a diagram illustrating a schematic configuration of each of a terminal, a host, and a server in the system configuration diagram illustrated in FIG.
FIG. 3 is a diagram illustrating each component in the terminal shown in FIG. 2 in more detail.
4 is a flowchart illustrating self-diagnosis start signal detection processing in the self-diagnosis start signal detection unit shown in FIG. 3;
[Explanation of symbols]
T terminal
H Host (host computer)
N network
S server
1 Voice input processor
2 Self-diagnosis department
3 Audio output processing section
4 Wireless communication unit (terminal T side)
11 Microphone
12 Voice input part
13 Voice temporary storage
14 Voice segment detector
21 Self-diagnosis start signal detector
22 Self-diagnosis department
23 Error message storage
31 Audio output unit
32 headphones
41 Signal transmitter
42 Signal receiver
61 Wireless communication unit (host computer H side)
62 Voice recognition unit
63 Application Department
64 Speech synthesis unit
65 Network communication unit (host computer H side)
71 Network communication unit (server S side)
72 Database

Claims (10)

  The user inputs a voice command from a terminal possessed by the user, the voice command input from the terminal is recognized by a host computer, and a response to the recognition result of the voice command is transmitted to the terminal. When there is a response abnormality in the response, a voice information providing system having means for notifying the user of the cause of the response abnormality, and when there is a response abnormality in the response, the cause of the response abnormality is The means for notifying the user includes a self-diagnosis start signal detection means for obtaining a self-diagnosis start signal by detecting an operation performed by the user when the user senses the response abnormality, and the self-diagnosis start Self-diagnosis means for starting self-diagnosis in response to the instruction signal and self-diagnosis results obtained by the self-diagnosis means are used. The operation performed by the user necessary for obtaining the self-diagnosis start signal is that the user confirms the operation state of the microphone with respect to the microphone present in the terminal. A voice information providing system characterized by operation.   2. The voice information providing system according to claim 1, wherein both a response from the host computer side to the voice command and a notification of the self-diagnosis result to the user are performed by voice.   The operation for confirming the operation state of the microphone is at least one of an operation in which the user blows into the microphone, an operation in which the user inputs predetermined sound into the microphone, and an operation in which the user taps the microphone. 3. A voice information providing system according to claim 1, wherein the information providing system is based on voice.   4. The voice information providing system according to claim 1, wherein the self-diagnosis start signal detection unit, the self-diagnosis unit, and the notification unit are provided mainly on the terminal side.   5. The voice information according to claim 1, wherein when there is a response abnormality in a response from the host computer to the voice command, the voice command input by the user is fed back to the user. Offer system.   The user inputs a voice command from a terminal possessed by the user, the voice command input from the terminal is recognized by a host computer, and a response to the recognition result of the voice command is transmitted to the terminal. When there is a response abnormality in the response, the malfunction cause notification method of the information providing system by voice having means for notifying the user of the cause of the response abnormality, and when there is a response abnormality in the response, The means for notifying the user of the cause of the response abnormality is to obtain a self-diagnosis start motion signal by detecting an operation performed by the user when the user detects the response abnormality on the host computer side. The self-diagnosis function is activated by the self-diagnosis start signal, and self-diagnosis is performed to investigate the cause of the response abnormality. The operation performed by the user necessary for obtaining the self-diagnosis start signal is the operation state of the microphone relative to the microphone present in the terminal. A malfunction cause notification method of an information providing system by voice, which is an operation to confirm.   7. The method of notifying a cause of malfunction of an information providing system by voice according to claim 6, wherein a response from the host computer to the voice command and a notification of the self-diagnosis result to the user are both performed by voice.   The operation for confirming the operation state of the microphone is at least one of an operation in which the user blows into the microphone, an operation in which the user inputs predetermined sound into the microphone, and an operation in which the user taps the microphone. 8. The malfunction cause notification method for an information providing system by voice according to claim 6.   9. The function of obtaining the self-diagnosis start signal, the function of performing the self-diagnosis, and the function of notifying the user of a self-diagnosis result are mainly provided on the terminal side. Of cause of malfunction of information providing system by voice.   10. The information provision by voice according to claim 6, wherein when a response from the host computer to the voice command is abnormal, the voice command inputted by the user is fed back to the user. System malfunction cause notification method.

Images