JPH10133849A - Personal computer and method for error notification - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a personal computer capable of ensuring a suitable user interface by exhibiting its reason when a voice recognition is failed. SOLUTION: A voice command utility program 31, when a voice signal of a user if stored in a buffer provided in a sound device driver 37, makes a voice recognizing engine 34 extract a word most suitable to that stored voice signal, learns a command corresponding to this extracted word referring to a voice command table 32, and issues this leant command. On the other hand, the voice command utility program 31, if the voice recognizing engine 34 fails a voice recognition processing and returns a return code showing its reason, learns an error message corresponding to the return code referring to an error message table 33, and requests displaying this learnt error message to an operating system 30.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a personal computer having a voice recognition function and supporting input of commands by voice, and an error notification method. The present invention relates to a personal computer and an error notification method for ensuring an appropriate user interface by presenting a user with an error notification.

[0002]

2. Description of the Related Art In recent years, the spread of computers has been remarkable, and various personal computers have been developed with this spread. In order to improve the operability of these personal computers, not only a character-based interface such as inputting commands from a keyboard, but also a picture displayed on a display,
A GUI (Graphic) that provides an interface by operating a pointing device such as a mouse, a pointing stick, and a trackball.
cal user interface) is becoming mainstream.

Recently, a speech recognition system has been studied as a man-machine interface for various devices, and a personal computer for inputting commands using speech has been put to practical use. As a result, the user can literally operate the personal computer through dialogue, and the operability has been dramatically improved. The command input function by voice is provided as an auxiliary function in the system, and it is not always necessary to input all commands using voice. Therefore, for example, only relatively frequently used operation instructions such as start and end of an application program are performed using voice, while for complicated operation instructions, a command is input from a keyboard or a pointing device is used as before. By performing different uses, such as by using the system, it is possible to prevent the scale of the system from becoming unnecessarily large.

However, in a personal computer to which a conventional speech recognition system is applied, the result is displayed only when the recognition processing for the inputted speech is successful,
The mainstream was to employ an interface to prompt the user for confirmation. Therefore, if the recognition process fails, the user does not know why the intended operation is not performed, so that there has been a problem that, for example, voice input in an incorrect environment is repeated.

[0005]

As described above, in a conventional personal computer that supports input of a command by voice, the result is displayed only when recognition processing of the input voice is successful, and the user is prompted for confirmation. The mainstream of adopting such an interface is that there is a problem that the user repeats voice input in an incorrect environment.

The present invention has been made in view of such circumstances, and when a recognition process for input speech fails, the reason is presented to the user to secure an appropriate user interface. It is an object to provide a computer and an error notification method.

[0007]

A personal computer according to the present invention recognizes an input voice and extracts a suitable phrase, and issues a command corresponding to the phrase extracted by the voice recognition unit. A personal computer having command issuing means, wherein the voice recognizing means includes means for returning a value indicating the reason as a return value when the voice recognition has failed, A message table that manages the character string in association with the character string representing the reason indicated by the value, and when the voice recognition unit returns any value as a return value,
Display means for acquiring a character string corresponding to the return value from the message table and displaying the acquired character string on a screen.

In the personal computer of the present invention, when the voice recognition means fails in voice recognition,
Return the value indicating the reason as the return value. Also, these return values and a character string expressing the reason indicated by the return value are registered in advance in the message table in association with each other. For example, "speak louder", "microphone input level is too high", "surrounding noise is too loud", and "not recognized as voice". Then, when the voice recognition means fails in voice recognition, that is, when a value indicating the reason is returned as a return value, the display means obtains a character string corresponding to the return value from the message table. Then, the obtained character string is displayed on the screen.

[0009] With this, when the user does not perform the intended operation in spite of the utterance, the user can quickly know the reason, and the voice input in an incorrect environment is repeated. This can be prevented, and an appropriate user interface can be secured.

Further, the personal computer of the present invention has a voice recognition means for recognizing an input voice and extracting a suitable phrase, and a command issuing means for issuing a command corresponding to the phrase extracted by the voice recognition means. Wherein the voice recognition means includes means for returning a value indicating the reason as a return value when the voice recognition has failed, and is represented by the return value and the return value. A message table that manages the voice in association with the voice expressing the reason, and when the voice recognition unit returns any value as a return value, obtains a voice corresponding to the return value from the message table. ,
Voice output means for outputting this voice.

[0011] In the personal computer of the present invention, instead of managing a character string expressing the reason for voice recognition failure, voice (voice waveform signal) expressing the reason is managed in a message table. Then, when the voice recognition means returns a value indicating the reason as a return value,
The sound output means obtains a sound corresponding to the return value, and outputs the obtained sound from a speaker or the like.

That is, the user can know the reason for the failure of speech recognition from the output voice, and can prevent the voice input in an incorrect environment from being repeated in the same manner as described above. Interface reservation is realized.

If, for example, a voice synthesizing means for synthesizing voice from a character string is provided, the message table can be managed by a character string and a notification to the user can be made by voice. Such maintenance work can be easily performed. In addition, it is possible to easily display the character string to the user together with the voice notification.

[0014]

BEST MODE FOR CARRYING OUT THE INVENTION

(First Embodiment) Hereinafter, a first embodiment of the present invention will be described with reference to the drawings.
An embodiment will be described. FIG. 1 is a diagram showing the device configuration of the personal computer according to the embodiment.

As shown in FIG. 1, the portable information device of the embodiment comprises a CPU 11, a system controller 12,
The system includes a system memory 13, a BIOS-ROM 14, a keyboard controller (KBC) 15, a sound controller 18, a magnetic disk device (HDD) 20, a display controller 21, and an I / O controller 25.

The CPU 11 controls execution of application programs including an operating system and utilities stored in the system memory 13. The system controller 12 controls the memory and I / O in this system.
This is a gate array for controlling O. The system memory 13 stores an operating system executed and controlled by the CPU 11, an application program including utilities, user data created by the application program, and the like.

The BIOS-ROM 14 has a system-BI
This is for storing an OS and a VGA-BIOS, and is constituted by a flash memory so that a program can be rewritten.

The keyboard controller 15 is for controlling a keyboard (KB) 16 and a mouse 17. The keyboard controller 15 scans a key matrix of the keyboard 16, receives a signal corresponding to a pressed key, and converts the signal into a predetermined key code. (Scan code). The keyboard controller 15 has a function of controlling the mouse 17 using a dedicated processor.

The sound controller 18 includes a PCM sound source and a digital signal processing circuit for audio signals.
The sound controller 18 is connected to a microphone input terminal to which the microphone 19 is connected, a line input terminal, a line output terminal, and a headphone terminal. Therefore, the voice uttered by the user is taken into the system via the sound controller 18.

The magnetic disk drive (HDD) 20 has a large capacity 2 for transmitting and receiving data to and from the system memory 13.
The next storage device. It should be noted that a table for associating a command string with a phrase, which is a feature of the present invention, is secured on the magnetic disk device 20 or in the system memory 13 described above. The display controller 21 includes an LCD 23 and a CRT 2
4 for display control, receives display data from the CPU 11 and writes it into the VRAM 22. Then, the I / O controller 25 includes a speaker 26
In addition, it controls various inputs and outputs to a serial port and a printer port.

FIG. 2 shows functional blocks of the personal computer according to the embodiment having such a device configuration. As shown in FIG. 2, in the personal computer of the embodiment, the operating system 30 controls the entire system. Then, under the control of the operating system 30, the voice command utility program 31, the voice recognition engine 34, and the sound device driver 37 operate respectively.

Voice command utility program 3
1 is a program for issuing a command in accordance with a user's voice captured via the microphone 19. The voice command utility program 31 includes a voice command table 32 used to determine which command should be issued for a user's voice, and an error code returned when the voice recognition engine 34 fails in voice recognition. It operates using an error message table 33 managed in association with the error message expressing the reason indicated by the error code. FIG. 3 shows the storage format of the voice command table 32, and FIG. 4 shows the storage format of the error message table 33.

Voice command utility program 3
1 detects a voice input at a level exceeding a preset threshold value by monitoring a buffer provided in the sound device driver 37, and executes a voice command issuing process. The sound device driver 37 drives and controls the sound controller 18 and includes a microphone 19 and a buffer that circulates and accumulates the audio signal captured via the sound controller 18.

Voice command utility program 3
When detecting a voice input at a level exceeding a preset threshold value, 1 transfers the address in the buffer in which the detected voice signal is held to the voice recognition engine 34. The speech recognition engine 34 that has received this address refers to the standard pattern speech dictionary 36 to extract the word (reading) that best matches the speech signal indicated by that address. Then, the voice recognition engine 34
Searches the grammar table 35 using the extracted word (reading) as a search key, detects a command name corresponding to the extracted word, and returns it to the voice command utility program 31. The grammar table 35 is a table for high-speed search processing created based on the voice command table 32, and includes at least the correspondence between words (readings) and command names.

The voice command utility program 31 that has returned the command name searches the voice command table 32 using the command name as a search key.
The command (action) associated with the command name is issued.

On the other hand, if the speech recognition engine 34 cannot extract a phrase matching the speech for some reason, it returns an error code indicating the reason to the speech command utility program 31. At this time,
The voice command utility program 31 uses this error code as a search key to
And requests the operating system 30 to display on the screen an error message associated with the error code.

Thus, the user can quickly know the reason for the failure of the speech recognition processing, so that an appropriate user interface can be secured. Next, an operation procedure of the personal computer of the embodiment will be described with reference to FIG.

Voice command utility program 3
1 detects an input of a voice (step A1), the voice recognition engine 34 extracts a phrase that best matches the input voice (step A2). If the extraction is successful (Y in step A3), the command name corresponding to the extracted phrase is returned to the voice command utility program 31. Then, the voice command utility program 31 searches the voice command table 32 using the command associated with the command name as a search key (step A4), and issues the command associated with the command name (step A4). A5).

On the other hand, if this extraction fails (N in step A3), the voice command utility program 3
1 receives an error code indicating the reason from the speech recognition engine 34 (step A6), and uses the received error code as a search key to generate an error message table 33.
(Step A7). Then, it requests the operating system 30 to display an error message associated with the error code on the screen (step A8).

As described above, by displaying the reason for the failure of the voice recognition processing, it is possible to prevent the voice input in an incorrect environment from being repeated. (Second Embodiment) Next, a second embodiment of the present invention will be described. The device configuration of the personal computer according to the second embodiment is the same as that of the first embodiment.

FIG. 6 shows functional blocks of the personal computer of the embodiment. As shown in FIG. 6, the personal computer according to the second embodiment has a configuration in which a speech synthesis engine 38 is added to the first embodiment. When a character string is delivered, the voice synthesis engine 38 uses the standard pattern voice dictionary 36 to synthesize a voice from the character string. When the error code is returned from the voice recognition engine 34, the voice command utility program 31 searches the error message table 33 using the error code as a search key, and then searches the error message associated with the error code. Deliver to the speech synthesis engine 38.

The speech synthesis engine 38 which has received the error message refers to the standard pattern speech dictionary 36 to synthesize speech from the character string constituting the error message. Then, the voice command utility program 31 requests the operating system 30 to output the voice synthesized by the voice synthesis engine 38.

Thus, the user can know the reason for the failure of the voice recognition by voice instead of the screen display of the first embodiment. At this time, the display of the error message on the screen is simultaneously performed by the operating system 3.
You may request 0. According to this configuration, since the error message itself is managed as a character string, it is possible to easily perform a maintenance operation such as changing the content while realizing an error notification function by voice.

Further, even in the configuration without the voice synthesis engine 38, for example, the error message table 33
Is configured as shown in FIG. 7, it is possible to execute notification of the reason for the failure of voice recognition by voice. That is, instead of managing the error message by a character string, a voice waveform of a voice expressing the reason for the failure is managed in advance. As a result, notification of the reason for the failure of voice recognition by voice is realized without increasing the scale of the system.

The error notification method described in the first embodiment is a program that can be executed by a computer, such as a magnetic disk such as a floppy disk or a hard disk, an optical disk such as a CD-ROM or a DVD, or a recording medium such as a semiconductor memory. It can be stored and distributed on a medium.

[0036]

As described above in detail, according to the present invention, when the intended operation is not performed despite the utterance, the reason is notified by screen display or voice output. The user can get to know the information promptly, and it is possible to prevent the voice input from being repeated in an incorrect environment, and to secure an appropriate user interface.

[Brief description of the drawings]

FIG. 1 is an exemplary view showing a device configuration of a personal computer according to first and second embodiments of the present invention.

FIG. 2 is an exemplary view showing functional blocks of the personal computer according to the first embodiment;

FIG. 3 is a conceptual diagram showing a storage format of a voice command table according to the first embodiment.

FIG. 4 is a conceptual diagram showing a storage format of an error message table according to the first embodiment.

FIG. 5 is an exemplary flowchart for explaining the operation procedure of the personal computer according to the first embodiment;

FIG. 6 is an exemplary view showing functional blocks of a personal computer according to a second embodiment;

FIG. 7 is a conceptual diagram showing a storage format of an error message table according to the second embodiment.

[Explanation of symbols]

11 CPU, 12 system controller, 13 system memory, 14 BIOS-ROM, 15 keyboard controller (KBC), 16 keyboard (K
B), 17: mouse, 18: sound controller, 1
9 microphone, 20 magnetic disk device (HDD), 21
... display controller, 22 ... VRAM, 23 ... LCD,
24 CRT, 25 I / O controller, 26 speaker, 30 operating system, 31 voice command utility program, 32 voice command table, 33 error message table, 34 voice recognition engine, 35 grammar table, 36: Standard pattern voice dictionary, 37: Sound device driver,
38 ... Speech synthesis engine.

Claims (8)

[Claims] 1. A personal computer comprising: voice recognition means for recognizing an input voice and extracting a suitable phrase; and command issuing means for issuing a command corresponding to the phrase extracted by the voice recognition means. In the above, the voice recognition means, when the voice recognition failed, comprising means for returning a value indicating the reason as a return value, the return value and a character string expressing the reason indicated by the return value And a message table that manages them in association with each other. When the voice recognition unit returns any value as a return value, a character string corresponding to the return value is obtained from the message table, and the information is obtained. A personal computer, comprising: display means for displaying a character string on a screen. 2. A personal computer comprising: voice recognition means for recognizing an input voice and extracting a suitable phrase; and command issuing means for issuing a command corresponding to the phrase extracted by the voice recognition means. In the above, the voice recognition means, when the recognition of the voice has failed, comprising means for returning a value indicating the reason as a return value, the return value and the voice expressing the reason indicated by the return value A message table that is managed in association with a voice that, when the voice recognition unit returns any value as a return value, obtains a voice corresponding to the return value from the message table and outputs the voice. A personal computer comprising output means. 3. A personal computer comprising: voice recognition means for recognizing an input voice and extracting a matching phrase; and command issuing means for issuing a command corresponding to the phrase extracted by the voice recognition means. In the above, the voice recognition means, when the voice recognition failed, comprising means for returning a value indicating the reason as a return value, the return value and a character string expressing the reason indicated by the return value A message table that manages the character strings in association with each other, a voice synthesis unit that synthesizes voice from the input character string, and a character corresponding to the return value when the voice recognition unit returns any value as a return value. A voice output unit that obtains a sequence from the message table, causes the voice synthesis unit to synthesize a voice that matches the obtained character string, and outputs the synthesized voice. A personal computer, characterized by comprising comprises a. 4. A personal computer comprising: voice recognition means for recognizing an input voice and extracting a suitable phrase; and command issuing means for issuing a command corresponding to the phrase extracted by the voice recognition means. In the above, the voice recognition means, when the voice recognition failed, comprising means for returning a value indicating the reason as a return value, the return value and a character string expressing the reason indicated by the return value A message table that manages the character strings in association with each other, a voice synthesis unit that synthesizes voice from the input character string, and a character corresponding to the return value when the voice recognition unit returns any value as a return value. A display unit that obtains a column from the message table and displays the obtained character string on a screen; and the voice recognition unit returns any value as a return value. At this time, a character string corresponding to the return value is obtained from the message table, a voice matching the obtained character string is synthesized by the voice synthesis means, and a voice output means for outputting the synthesized voice is provided. A personal computer characterized by comprising: 5. A speech recognition device comprising: speech recognition means for recognizing an input speech and extracting a matching phrase; and command issuing means for issuing a command corresponding to the phrase extracted by the speech recognition means. A means for returning a value indicating the reason as a return value when a failure has occurred in the voice recognition means, and managing the return value and a character string expressing the reason indicated by the return value in association with each other. An error notification method for a personal computer comprising a table, wherein when the voice recognition means returns any value as a return value, a character string corresponding to the return value is obtained from the message table. An error notification method characterized by displaying a learned character string on a screen. 6. A speech recognition device comprising: speech recognition means for recognizing an input speech and extracting a suitable phrase; and command issuing means for issuing a command corresponding to the phrase extracted by the speech recognition means. A message table for providing a means for returning a value indicating the reason as a return value when a failure has occurred in the voice recognition means, and managing the return value and a voice expressing the reason indicated by the return value in association with each other An error notification method for a personal computer, comprising: when the voice recognition unit returns any value as a return value, obtains a voice corresponding to the return value from the message table; An error notification method, characterized in that it outputs an output sound. 7. A speech recognition apparatus comprising: speech recognition means for recognizing input speech and extracting a suitable phrase; and command issuing means for issuing a command corresponding to the phrase extracted by the speech recognition means. A means for returning a value indicating the reason as a return value when a failure has occurred in the voice recognition means, and managing the return value and a character string expressing the reason indicated by the return value in association with each other. An error notification method for a personal computer comprising a table, wherein when the voice recognition means returns any value as a return value, a character string corresponding to the return value is obtained from the message table. An error notification method, comprising: synthesizing a voice from a known character string; and outputting the synthesized voice. 8. A speech recognition device comprising: speech recognition means for recognizing an input speech and extracting a suitable phrase; and command issuing means for issuing a command corresponding to the phrase extracted by the speech recognition means. A means for returning a value indicating the reason as a return value when a failure has occurred in the voice recognition means, and managing the return value and a character string expressing the reason indicated by the return value in association with each other. A program for notifying an error on a personal computer having a table, wherein when the voice recognition means returns any value as a return value, a character string corresponding to the return value is known from the message table. A computer-readable storage medium storing a program for controlling a computer to display the obtained character string on a screen.