JP2002268683A - Method and device for information processing - Google Patents

Abstract

PROBLEM TO BE SOLVED: To actualize stable voice recognition which is not affected by environmental changes by making it possible to properly set operation conditions of voice recognition processing corresponding to the environmental changes. SOLUTION: A voice recognition part is provided which inputs a vocal indication to a presentation application being executed. According to words inputted from the voice recognition part, and environmental noise and lightness, the current progress state of an assumed schedule is discriminated (step S410). According to the discriminated progress state, the operation conditions of the voice recognition part are set (steps S412 to S415).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information processing method and apparatus which can be operated by voice, and more particularly to an information processing method and apparatus suitable for presenting materials in a lecture such as a presentation.

[0002]

2. Description of the Related Art Conventionally, in a presentation, a presenter has performed presentation while manually replacing presentation materials printed on an OHP sheet. Also, recently, it has become common to create presentation data using presentation software on a personal computer (PC) device and present presentation materials using a PC device and a projector.

[0003] This kind of presentation software is PC
Although it is common to operate using the keyboard and mouse of the device, some methods of operating presentation software on a PC device using voice recognition have been proposed (for example, Japanese Patent No. 0924717, Japanese Patent Application Laid-Open No. 6-31
No. 8235). The advantage of operation using voice recognition is that pointing operations with a mouse or the like, which are necessary when operating presentation software such as turning pages, outputting sound effects, and displaying animations, or keyboard operations can be omitted. You can concentrate on the presentation. In addition, there is an advantage that the presentation is not interrupted by the operation, and the presenter does not necessarily need to be in front of the PC, and the presentation style can be varied.

[0004]

However, the accuracy of speech recognition is susceptible to changes in the environment in which it is used, such as differences in speakers, changes in ambient noise, and the like. Optimum settings for speech recognition operating conditions such as search conditions differ depending on the environment in which they are used.

[0005] For example, in a general presentation, a flow generally starts with an introduction of a lecture by a moderator, a lecture by the lecturer, and a question-and-answer session after the lecture.
In these flows, the environment related to speech recognition changes depending on the progress of the lecture. For example, taking a speaker as an example, a speaker during a lecture is often limited to the speaker, but in a question-and-answer session, there are more opportunities for an auditor and a moderator to speak in addition to the speaker. In addition, the noise in the venue, which is a noise for voice recognition, changes according to the progress of the lecture. For example, in the case of a general lecture, before and after the lecture, the listeners are busy with each other's voice,
During the lecture, the attention of the listeners is transferred to the lecture, and the noise is reduced.

[0006] Therefore, the accuracy of voice recognition is affected by changes in the environment that occur in the flow of presentation as described above, and there is a possibility that stable operations cannot be performed. Conventionally proposed methods use speech recognition to operate a presentation,
None of the proposals suggests changing the operating conditions of the speech recognition process in order to improve the accuracy of speech recognition during the presentation.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems, and enables an operation condition of a voice recognition process to be appropriately set according to a change in an environment, thereby realizing stable voice recognition regardless of a change in the environment. It is an object of the present invention to provide an information processing apparatus and method for performing the above.

Another object of the present invention is to detect the progress of a schedule, such as a lecture, which is assumed in advance, and to set the operating conditions of the speech recognition process so as to be suitable for an environment assumed according to the progress. That is, it is possible to switch.

[0009]

An information processing apparatus according to the present invention for achieving the above object has, for example, the following arrangement. That is, based on a voice recognition unit for inputting an instruction to a running application by voice, a determination unit for determining a current progression stage in a previously assumed schedule, and a progression stage determined by the determination unit. Setting means for setting operating conditions in the voice recognition means.

According to another aspect of the present invention, there is provided an information processing method comprising: a voice recognition step for inputting an instruction to a running application by voice; and a current progress stage in a previously assumed schedule. And a setting step of setting operating conditions in the voice recognition step based on the progression stage determined in the determination step.

[0011]

Preferred embodiments of the present invention will be described below with reference to the accompanying drawings.

In the following embodiment, a presentation device that can be operated by voice, for example, which determines the progress of a presentation such as a presentation and changes the operation conditions of voice recognition according to the progress will be described.

As described above, in order to maintain stable recognition accuracy with respect to an environment change accompanying the progress of a presentation, it is necessary to select an operation condition suitable for the environment according to the environment change. For example, when moving from a quiet environment to an environment with high ambient noise, a common method is
It is necessary to adapt the acoustic model to noise and to perform noise removal processing represented by spectral subtraction at the time of acoustic analysis. Therefore, in a situation where the noise of the lecture hall is increased in the course of the presentation in the course of the presentation, as described above, the acoustic model is adapted to the noise or the acoustic analysis method is adapted to the noise environment. It is desirable to change to.

For example, for a change in the speaker,
During the lecture, an acoustic model (specific speaker model) created from the speech of the speaker was used, and at the time of the question and answer session, an acoustic model (unspecified speaker) created from the speech of many speakers so that speech other than the speaker could be recognized Model).

Hereinafter, the presentation apparatus of the present embodiment will be described in detail.

FIG. 1 is a block diagram showing an example of the configuration of a presentation device according to the present invention. In FIG. 1, the presentation device includes a display device 101 such as a display or a projector, a microphone, and the like.
A voice input device 102 used to input voice, a light intensity sensor 103 for detecting light intensity, an input device 104 including a pointing device such as a mouse and a keyboard, a central processing unit for processing electronic data such as a CPU 105, a storage device 10 such as a magnetic storage device or a memory
6 is provided.

FIG. 2 is a block diagram showing a functional configuration of the presentation device according to the present embodiment. Reference numeral 201 denotes a control unit that controls processing of the entire presentation apparatus. Reference numeral 202 denotes a voice detection unit which performs voice section detection. Reference numeral 203 denotes a speech recognition unit, which is a language model 2
The voice detection unit 202 recognizes the voice detected by the voice detection unit 202 using the audio model 14 and the acoustic model 213, and outputs the recognition result. The acoustic model 213 stores an acoustic model used for speech recognition such as an HMM, and the language model 214 stores a language model such as a word dictionary used for speech recognition.

Reference numeral 204 denotes a progress status judging unit, which detects a change in the progress status of the presentation based on the voice recognition result from the voice recognition unit 203 and a signal from the light amount sensor 103, and judges the progress status. The progress situation determination uses a progress situation determination table 211 that describes a condition for detecting a change in the progress of the presentation and a change in the progress of the presentation corresponding to the condition.

An operation condition selection unit 205 selects operation conditions for speech recognition corresponding to the progress of the presentation. An operation condition application unit 206 applies the selected operation condition to the speech detection unit 202 and the speech recognition unit 203. The operation condition selection unit 205 and the operation condition application unit 2
Reference numeral 06 refers to an operation condition table 212 that describes operation conditions for speech recognition corresponding to the progress of the presentation, and selects and applies the operation conditions.

Reference numeral 207 denotes a progress status determination table generation unit which generates a progress status determination table. An operation condition table creation unit 208 creates an operation condition table. Reference numeral 209 denotes an operation environment adaptation unit that adapts to an environment using an acoustic model, a threshold for detecting a speech section, and the like. 21
Reference numeral 0 denotes a presentation control unit, which controls presentation operations such as turning pages. Reference numeral 215 denotes presentation data, including data to be displayed on the display device 101 in the presentation.

Each of the functional units 201 to 210 is realized by the central processing unit 105 executing a control program stored in the storage device 106. Data indicated by 211 to 215 referred to by each function is stored in the storage device 106. That is, in a series of processing described below, basically, a program describing each function and data referred to by each function are loaded on the storage device (106), and the loaded program is loaded by the central processing unit (105). It is realized by executing.

The operation of this embodiment having the above configuration will be described with reference to flowcharts.

FIG. 3 is a flowchart showing an outline of the operation of the presentation device using the speech recognition according to the present embodiment.

First, in step S300, the operating conditions for the voice recognition processing are updated. This is because the progress determination unit 204 determines the progress of the presentation, and the operation condition selection unit 205 and the operation condition application unit 206
Is to set operating conditions according to the determined progress in the voice detection unit 202 and the voice recognition unit 203.
Details of this processing will be described later with reference to FIGS. 4A and 4B. In step 301, the utterance of the user of the presentation device is fetched via the voice input device 102 such as a microphone, and the voice section is detected by the voice detection unit 201. The voice detection unit 201 divides the voice data input from the voice input device 102 into blocks for each unit time (for example, 0.01 seconds), and the voice data is a voice section or a non-voice section for each block. Has been determined. If it is detected by the voice detection unit 202 as a voice section, the process proceeds to step S302, and the voice recognition unit 203 performs voice recognition on the captured voice.

If the recognition result is a specific keyword for giving an instruction to the presentation device, for example, "next page" or "last page", the process proceeds from step S303 to step S304, and according to the recognition result. The presentation control is performed by the presentation control unit 210 (steps 304 to 30).
5). For example, when the keyword “next page” is recognized, control is performed to display the presentation data of the next page on the display device 101. If the recognition result indicates termination, the process is terminated. On the other hand, when a word other than the specific keyword is recognized in step 303, the process returns to step S300, and the processing for updating the operation conditions of voice recognition and the processing for detecting a voice section are performed again.

A description will now be given of a control of detecting the progress of the presentation and switching the operating conditions of the speech recognition processing in accordance with the progress of the presentation in the presentation apparatus of the present embodiment that operates as described above.

Next, a description will be given of the processing for changing the operating conditions for speech recognition in step S300 described above. Note that this operation condition change processing is performed by the progress status determination unit 204,
The operation condition selection unit 205 and the operation condition application unit 206 perform the operation with reference to the progress status determination table 211 and the operation condition table 212. Therefore, the user needs to set the progress judgment table 211 and the operation condition table 212 prior to the presentation. First, a table setting operation will be described.

FIG. 4A is a flowchart illustrating a table setting process according to the embodiment. If a predetermined operation for table setting processing is input from the input device 104 and the operation instructs the creation processing of the progress judgment table, the process proceeds from step S401 to step S402, and the progress judgment table creation unit 207 Is started, and a screen for creating a progress judgment table is displayed on the display device 101. Then, in step S 403, various operation inputs to the creation screen are accepted, a progress status determination table 211 is created, and this is stored in the storage device 106.

The progress judgment table 211 created by the progress judgment table creation unit 207 describes conditions for detecting a change in the progress and the progress corresponding to the condition. Here, the progress status is, for example, “before the lecture”
Shows the progress and status of presentations such as "during lecture", "question and answer", and "end of lecture", and can be freely defined by the user. For detecting the change in the progress, information such as a result of the immediately preceding speech recognition, a change in the amount of light in the lecture hall, and a change in the noise level are used. Then, based on the progress of the presentation at that time and the above information, it is determined whether or not the progress of the presentation is close.

FIG. 5 shows an example of the progress judgment table. FIG. 5 is an example of a case where a change in the progress of the presentation is detected when a specific keyword is recognized by the immediately preceding voice recognition unit. When the progress status is "Before Lecture" and the speech "Start", "Begin", or "Start", which means the start of the lecture, is recognized, it is determined that the presentation will be started, and the process proceeds. It is described that the situation is "in lecture". In addition, if the progress status is "in lecture" and the speech "finished" or "it is over" is recognized, it is determined that the presentation has entered the "question and answer" stage, and the progress status is determined. Is "Question and Answer". In addition, as shown in FIG.
A plurality of conditions (here, recognition results) for detecting a change in the progress status are associated with one change in the progress status in the progress status determination table.

Similarly, FIG. 6 is an example of a progress judgment table for judging the progress of a presentation based on a change in the brightness of the illumination of the lecture hall. The brightness of the illumination in the lecture hall, that is, the light amount is detected by the light amount sensor 103. In this example, it is assumed that the illumination in the lecture hall of “during lecture” is darkened, and conversely, the illumination is brightened again at “question and answer” and “end of lecture”. Therefore, in the progress status determination table of FIG. 6, the current progress status is “before lecture”.
If the light amount falls below a certain value, the result of the determination of the progress is regarded as “in lecture”, and if the current progress is “in lecture” and the light amount exceeds a certain value, the progress is determined. Is described as "question and answer".

Similarly, FIG. 7 shows an example of the progress status judgment table when the progress status is judged based on the change in the noise level. The “noise level” referred to here means a non-speech section (a section where the user does not emit a sound), that is, a loudness of ambient noise such as a noise of a hall input through a microphone. In order to measure the loudness, it is necessary to perform processing of (1) separation between a voice section and a non-voice section, and (2) calculation of a loudness of a sound section in the non-voice section. The voice section and the non-voice section (1) are separated into voice and non-voice based on the result of the voice detection section 202. In general, amplitude and power are used as the loudness of the sound in the non-voice section of (2). In this example,
During the lecture, the inside of the lecture hall is relatively quiet, but it is assumed that it will be a little annoying during the question and answer session. Therefore, FIG.
In the progress judgment table of the above, if the current progress is “before lecture” and the noise level is lower than a predetermined threshold Th, the progress is switched to “during lecture”, and the current progress is changed to “during lecture”. Indicates that the progress status is switched to “question and answer” when the noise level is equal to or greater than the threshold value Th. The noise level is detected by the voice input device 1
02 is performed on the basis of the signal input through the input terminal 02.

Referring back to FIG. 4A, if a predetermined operation for table setting processing is input from the input device 104 and the operation instructs the creation of the operation condition table, the processing proceeds from step S401 to step S404, and returns to step S404. S
Proceed to 405. In step S405, the operation condition table creation unit 208 is activated, and a screen for creating an operation condition table is displayed on the display device 101. Then, step S4
At step 06, various operation inputs to the creation screen are accepted, an operation condition table 212 is created, and this is stored in the storage device 106.

The operating condition table 212 describes operating conditions for speech recognition corresponding to the progress. The items of the operating conditions include, for example, an acoustic model used for speech recognition,
Grammar or language model used in speech recognition, acoustic analysis method, various conditions when performing acoustic analysis, thresholds used for voice section detection, various search conditions used when performing voice recognition,
There are various threshold values used for rejection processing of the speech recognition result.

FIG. 8 shows an example of the operation condition table. FIG.
In the operating condition table, the acoustic model to be selected in each progress situation, the threshold for detecting the voice section, and the presence or absence of the environment adaptation process are described. As shown in FIG. 8, a plurality of operation conditions can be described for one progress state in the operation condition table.

When the end of the table creation processing is instructed from the input device 104, the processing is terminated from step S407.

Next, processing for changing the operating conditions of the speech recognition processing during the presentation, that is, step S30
The processing of 0 will be described. Before the detection of the voice section, the operation conditions for voice recognition are set using the progress status determination table 211 and the operation condition table 212 created by the processing of FIG. 4A.

First, in step S410, the progress judgment unit 204 judges the progress with reference to the progress judgment table 211. In the present embodiment, the current progress status, the recognition result from the voice detection unit 202 and the voice recognition unit 203, the light amount signal from the light amount sensor 103, and the voice input are used by using the progress state determination tables shown in FIGS. The progress is determined based on the noise level obtained based on the signal from the device 102. If the result of this determination is that the progress has changed, the process proceeds from step S411 to step S412. In step S412, the operation condition selection unit 205 refers to the operation condition table 212 and selects an operation condition for speech recognition corresponding to the determination result (progress) of the progress determination unit 204.

For example, using the progress judgment table shown in FIG. 5, when the progress is "in lecture", "it is over"
Is recognized, the progress status determining unit 204 determines that the progress status is “in lecture” according to the progress status determination table (FIG. 5).
Is changed to “Q & A” (Step 4)
10, S411). In this case, since the progress has changed, the operation condition selection unit 205 subsequently refers to the operation condition table and selects an operation condition corresponding to the progress “question and answer” determined in step S410. When the operation condition table of FIG. 8 is used, the acoustic model for the unspecified speaker is selected, and the threshold for detecting the voice section is changed to Th3.
At the same time, a condition for performing the environment adaptation process is selected.

Next, the process proceeds to step S413, in which various operating conditions for speech recognition performed thereafter are changed in accordance with the operating condition selected in step S412. At this time, if the operation condition for performing the environment adaptation process is selected, the environment adaptation process of the acoustic model is performed in the operation environment adaptation unit 209, and the acoustic model after the adaptation is stored in the storage device 106 (steps S414 and S415). ). The environment adaptation processing here is
It mainly refers to a method involving a change in an acoustic model, and indicates a technique such as CMS, MLLR, PMC, etc. as a known technique. Then, in the voice recognition processing after step S301, voice recognition is executed under the newly applied operating condition.

As described above, according to the above embodiment, the progress of the presentation is detected based on the content of the sound, the brightness of the venue, and the noise level, and the progress is preliminarily assigned to each progress according to the detected progress. The operating condition of the speech recognition process is set. Therefore, an appropriate voice recognition process can be realized according to the progress, and a stable presentation operation by voice can be performed.

In the above-described embodiment, all the tables shown in FIGS. 5 to 7 are used when the progress judgment is made by the progress judgment section 204, but a part of them may be used. For example, when the progress situation determination table shown in FIG. 6 is not used, the light amount sensor becomes unnecessary.

Further, in the progress status determination in step S410 of the above embodiment, when a specific operation of a pointing device such as a mouse or a specific key operation on a keyboard is performed, a change in the progress status is detected to detect the progress status. May be determined. FIG. 9 shows an example of the progress status determination table in this case. In this case, a key operation for switching the progress is assigned in advance and registered in the progress determination table.

Further, in the progress status determination in step S410 of the above embodiment, the determination may be made based on the execution status of the application (presentation program in this example). For example, detection of a change in the progress when specific data is displayed among presentation data used for a lecture can be cited. More specifically, for example, each display page included in the presentation data is given attributes such as a title portion, a first page,..., The last page, etc., and by referring to the progress judgment table shown in FIG. The progress is determined based on the attributes of the page. In addition, FIG.
Only one of the progress status determination tables as shown in FIGS. 7, 9, and 10 may be used, or a plurality of progress status determination tables may be used in combination.

In the operation condition selection in step S412 of the embodiment, the operation condition is automatically selected with reference to the operation condition table 212. However, a window as shown in FIG. 11 is displayed and a pointing device such as a mouse is displayed. Alternatively, the speaker may be allowed to directly select an operation condition using a keyboard. In this case, the display of the setting screen may be triggered by a change in the progress status, or may be displayed by a predetermined operation.

In addition, the progress determined in the progress determination in step S410 of the above embodiment may be notified to the speaker. Also, similarly, step S412
The operating condition selected in the operating condition selection may be notified to the speaker. Here, as a method of notifying the speaker, the determined progress or the selected operating condition may be displayed on the display device 101. In the device configuration of the present apparatus, a plurality of display devices may be provided, and the notification content may be displayed on a display device that can be viewed only by the speaker. FIG. 12 shows an image diagram when the image is displayed on the display device.

The progress status may be selected by operating the progress status portion displayed on the display device with a pointing device such as a mouse as shown in FIG. Similarly, the operating condition may be selected by operating a portion of the operating condition displayed on the display device with a pointing device such as a mouse. By doing so, when it becomes difficult to recognize the keyword during the presentation, the speaker can arbitrarily change the progress status or operating conditions even during the presentation.

Further, in the above embodiment, the items of the operation conditions that can be described in the operation condition table are not limited to those shown in FIG. 8, and include, for example, a grammar or language model used in speech recognition, an acoustic analysis method (MFCC analysis). , LPC-MEL analysis, S
pectral Subtraction, Cepstrum Mean Subtraction)
And various conditions for acoustic analysis (sampling frequency, analysis window width, mel coefficient, spectral subtraction coefficient, parameter type), and various conditions for language search (whether or not to perform detailed search) , The type of language model to be used, the weight of language likelihood, the processing reduction method and beam width used in the search, the number of candidates to be searched), various thresholds used for rejection processing of speech recognition results, and the presence or absence of speaker adaptation processing. It goes without saying that any operation conditions relating to the operation of voice recognition may be described.

As described above, according to the present embodiment,
By automatically judging the progress of the presentation and changing the operating conditions of speech recognition according to the progress,
There is an effect that a presentation device using voice recognition with high recognition accuracy can be realized regardless of the progress of the presentation.

Another object of the present invention is to provide a system or an apparatus with a storage medium storing software program codes for realizing the functions of the above-described embodiments,
And MPU) by reading and executing the program code stored in the storage medium.

In this case, the program code itself read from the storage medium implements the functions of the above-described embodiment, and the storage medium storing the program code constitutes the present invention.

As a storage medium for supplying the program code, for example, a floppy disk, hard disk, optical disk, magneto-optical disk, CD-ROM, CD
-R, a magnetic tape, a nonvolatile memory card, a ROM, or the like can be used.

When the computer executes the readout program code, not only the functions of the above-described embodiment are realized, but also the OS (Operating System) running on the computer based on the instruction of the program code. ) May perform some or all of the actual processing, and the processing may realize the functions of the above-described embodiments.

Further, after the program code read from the storage medium is written in a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer, the program code is read based on the instruction of the program code. It goes without saying that the CPU included in the function expansion board or the function expansion unit performs part or all of the actual processing, and the processing realizes the functions of the above-described embodiments.

[0055]

As described above, according to the present invention,
It is possible to appropriately set the operating conditions of the voice recognition process following changes in the environment, and to achieve stable voice recognition that is not affected by changes in the environment. Further, according to the present invention, it is possible to detect a progress state of a schedule assumed in advance, such as a lecture, and to switch an operation condition of a voice recognition process according to the progress state so as to be suitable for an assumed environment. Become.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a basic configuration of a presentation device according to an embodiment.

FIG. 2 is a block diagram illustrating a functional configuration of the presentation device according to the embodiment.

FIG. 3 is a flowchart showing a procedure of a presentation using speech recognition in the presentation device according to the embodiment;

FIG. 4A is a flowchart illustrating a table creation process in the presentation device according to the embodiment;

FIG. 4B is a flowchart showing a processing procedure in a case where an operation condition of voice recognition is changed according to the progress of the presentation in the presentation device according to the embodiment.

FIG. 5 is a diagram illustrating an example of a data configuration of a progress status determination table when determining a progress status when a specific keyword is recognized.

FIG. 6 is a diagram illustrating a data configuration example of a progress status determination table when the progress status is determined based on a change in illumination of a lecture hall.

FIG. 7 is a diagram illustrating a data configuration example of a progress status determination table when the progress status is determined based on a change in noise level.

FIG. 8 is a diagram illustrating a data configuration example of an operation condition table.

FIG. 9 is a diagram illustrating an example of a data configuration of a progress determination table when determining a progress when a specific operation of a pointing device such as a mouse or a specific keyboard operation is performed.

FIG. 10 is a diagram illustrating an example of a data configuration of a progress determination table when determining a progress when specific data is displayed among presentation data used for a lecture.

FIG. 11 is a diagram illustrating an example of a setting screen when a speaker directly selects an operation condition using a pointing device or a keyboard.

FIG. 12 is a diagram illustrating a state of displaying a progress status or an operation condition on a display device.

 ────────────────────────────────────────────────── ─── Continued on the front page (72) Inventor Hiroki Yamamoto 3-30-2 Shimomaruko, Ota-ku, Tokyo F-term in Canon Inc. (reference) 5D015 HH23 KK01 LL02 LL03 LL10

Claims (38)

[Claims] 1. A voice recognition unit for inputting an instruction to an application being executed by voice, a determination unit for determining a current progression stage in a previously assumed schedule, and a progression stage determined by the determination unit A setting unit for setting an operating condition in the voice recognition unit based on the information. 2. The method according to claim 1, wherein the setting unit sets an acoustic model, a grammar or a language model, sets an acoustic analysis method for analyzing a speech, sets a condition for performing an acoustic analysis, and sets a speech section. 2. The information processing apparatus according to claim 1, wherein at least one of setting a detection threshold, setting a search condition at the time of speech recognition, and setting a threshold used for rejection processing of the speech recognition result is performed. 3. An operation condition table which describes operation conditions of the voice recognition means corresponding to the progression stage of the schedule, wherein the setting means sets the operation condition corresponding to the current progression stage determined by the determination means. 3. The information processing apparatus according to claim 1, wherein the information is extracted from the operation condition table, and an operation condition of the voice recognition unit is set according to the extracted operation condition. 4. The information processing apparatus according to claim 3, further comprising an operation condition table creating unit that provides a user interface for creating the operation condition table. 5. The apparatus according to claim 1, wherein said determination means determines whether or not a signal taken from outside satisfies a predetermined transition condition, and determines a progress stage based on the determination result. An information processing apparatus according to claim 1. 6. The information processing apparatus according to claim 5, wherein the transition condition includes the presence of a predetermined keyword in a recognition result by the voice recognition unit. 7. The information processing apparatus according to claim 5, further comprising a detection unit configured to detect a light amount of an environment, wherein the transition condition includes a light amount detected by the detection unit. 8. The information processing apparatus according to claim 5, further comprising a measuring unit that measures a noise level of the environment, wherein the transition condition includes a noise level measured by the measuring unit. 9. The information processing apparatus according to claim 5, wherein the transition condition includes whether or not a predetermined operation signal is input by an operator. 10. The system according to claim 5, wherein the transition condition includes a progress status of an application being executed.
An information processing apparatus according to claim 1. 11. The information processing apparatus according to claim 10, wherein the application includes an image presentation function, and the progress of the application is an image being presented by the application. 12. The system according to claim 1, further comprising a determination table in which said transition condition is described for each progress stage in said schedule, wherein said determination means determines said progress stage by referring to said determination table. Items 5 to 11
An information processing device according to any one of the above. 13. The information processing apparatus according to claim 12, wherein a plurality of transition conditions can be described for one transition in the determination table. 14. The information processing apparatus according to claim 12, further comprising table creation means for providing a user interface for creating the determination table. 15. The apparatus according to claim 1, wherein the setting unit provides a user interface for setting an operation condition in the voice recognition unit when a progression stage determined by the determination unit changes. Item 2. The information processing device according to item 1. 16. The information processing apparatus according to claim 1, further comprising a notifying means for notifying a progress stage determined by said determining means and an operation condition set by said setting means. 17. The information processing apparatus according to claim 16, wherein the notification unit displays the content of the notification on a display screen, and the display also serves as a setting screen for operating conditions. 18. The information processing apparatus according to claim 1, further comprising an operation environment learning unit that adapts an operation condition of speech recognition to an environment based on the progression stage determined by the determination unit. 19. A voice recognition process for inputting an instruction to a running application by voice, a determination process of determining a current progress stage in a previously assumed schedule, and a progress stage determined in the determination process. A setting step of setting an operation condition in the voice recognition step based on the information processing method. 20. The setting step includes setting an acoustic model used in the speech recognition step, setting a grammar or a language model, setting an acoustic analysis method for analyzing a speech, setting conditions for performing an acoustic analysis, and a speech section. 20. The information processing method according to claim 19, wherein at least one of setting a detection threshold, setting a search condition at the time of speech recognition, and setting a threshold used for rejection processing of the speech recognition result is performed. 21. An operation condition table in which operation conditions of the voice recognition step corresponding to the advance stages of the schedule are described, wherein the setting step sets the operation conditions corresponding to the current advance stage determined in the determination step. 21. An operation condition extracted in the operation condition table, and an operation condition of the voice recognition step is set according to the extracted operation condition.
0. The information processing method according to 0. 22. The information processing method according to claim 21, further comprising an operation condition table creating step of providing a user interface for creating the operation condition table. 23. The method according to claim 19, wherein the determining step determines whether or not a signal fetched from outside satisfies a predetermined transition condition, and determines a proceeding stage based on the determination result. An information processing method according to claim 1. 24. The information processing method according to claim 23, wherein the transition condition includes the presence of a predetermined keyword in a recognition result in the voice recognition step. 25. The information processing method according to claim 23, further comprising a detecting step of detecting a light amount of an environment, wherein the transition condition includes a light amount detected in the detecting step. 26. The information processing method according to claim 23, further comprising a measurement step of measuring a noise level of the environment, wherein the transition condition includes a noise level measured in the measurement step. 27. The information processing method according to claim 23, wherein the transition condition includes whether or not a predetermined operation signal is input by an operator. 28. The transition condition according to claim 2, wherein the transition condition includes a progress status of an application being executed.
3. The information processing method according to 3. 29. The information processing method according to claim 28, wherein the application includes an image presentation function, and the progress of the application is an image being presented by the application. 30. The apparatus according to claim 30, further comprising: a determination table describing the transition condition for each progression stage in the schedule, wherein the determination step determines the progression stage with reference to the determination table. Items 23 to 2
10. The information processing method according to any one of items 9. 31. The information processing method according to claim 30, wherein a plurality of transition conditions can be described for one transition in the determination table. 32. The information processing method according to claim 30, further comprising a table creation step of providing a user interface for creating the determination table. 33. The setting step provides a user interface for setting an operation condition in the voice recognition step when a progress stage determined in the determination step is changed. Item 20. The information processing method according to Item 19. 34. The information processing method according to claim 19, further comprising a notification step of notifying a progress stage determined in the determination step and an operation condition set in the setting step. 35. The information processing method according to claim 34, wherein in the notifying step, the content of the notification is displayed on a display screen, and the display also serves as a setting screen for operating conditions. 36. The information processing method according to claim 1, further comprising an operation environment learning step of adapting an operation condition of speech recognition to an environment based on the progress stage determined in the determination step. 37. A computer program for implementing the information processing method according to claim 19 by a computer. 38. A storage medium for storing a computer program for implementing the information processing method according to claim 19 by a computer.