JP2000259387A - Voice recognition device, storage medium recording voice recognition processing program and voice recognition processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a voice recognition device with which a progress condition of a voice recognition work is easily grasped even when voice recognition is performed for plural files at a time. SOLUTION: This voice recognition device is equipped with a storage medium 3 for recording a voice file having sound data, an operation member 11 which is capable of selecting plural voice files, a voice recognizing part 4 which continually convert the sound data of the selected voice files into text data, and a control part 4 which displays the progress a situation of text data conversion processing by the voice recognizing part 4A at every voice file.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice recognition device, a recording medium storing a voice recognition processing program, and a voice recognition processing method.

[0002]

2. Description of the Related Art There has been proposed a voice recognition apparatus which receives voice data recorded by a voice recording / reproducing apparatus, recognizes voice for each file by voice recognition software, and converts the file into text. In such a voice recognition device, when there are a plurality of voice data files for which voice recognition is to be performed, one file is selected by voice recognition software, the voice is recognized, and then the next file is selected. As in the case of speech recognition, a method of sequentially performing speech recognition for each file has been adopted. Since such a voice recognition method requires a considerable amount of time to complete voice recognition for all files, a method of automating the above operation and converting a plurality of files to text by continuously performing voice recognition at once is considered. Have been.

[0003]

However, when speech recognition for a plurality of files is performed automatically,
Since it is not possible to know the gap between the voice recognition for one file and the voice recognition for the next file, there is a problem that it is difficult to grasp the progress of voice recognition such as which file is currently being recognized.

The present invention has been made in view of such a problem, and an object of the present invention is to facilitate the progress of a voice recognition operation even when a plurality of files are voice-recognized at once. It is another object of the present invention to provide a voice recognition device, a recording medium storing a voice recognition processing program, and a voice recognition processing method.

[0005]

In order to achieve the above object, a voice recognition apparatus according to a first aspect of the present invention includes a recording medium for recording a voice file having voice data and a plurality of voice files. Selecting means capable of performing the following processing; a voice recognizing means for continuously converting voice data of a plurality of voice files selected by the selecting means into text data; and a voice indicating progress of the text data conversion processing by the voice recognizing means. Display means for displaying each file.

[0006] A speech recognition apparatus according to a second aspect of the present invention comprises:
In the first invention, the display means includes: file names of the plurality of audio files selected by the selection means;
A progress status indicating whether the text data conversion processing for the audio file corresponding to the file name has not been processed, is being processed, or has been processed is displayed.

[0007] A speech recognition apparatus according to a third aspect of the present invention comprises:
In the first invention, the display means displays the file names of the plurality of audio files selected by the selection means, and erases the display of the file names of the audio files for which the text data conversion processing has been completed.

[0008] A speech recognition apparatus according to a fourth invention is characterized in that:
In any one of the first to third inventions, the total audio data amount of the plurality of audio files selected by the selection unit and the audio data amount of the audio file corresponding to the text data converted by the audio recognition unit. And a calculating means for calculating a ratio of the audio data amount to the total audio data amount, and the display means further displays a calculation result of the calculating means.

According to a fifth aspect of the present invention, there is provided a recording medium storing a processing program for performing a voice recognition process by a computer, wherein the processing program stores a voice file having voice data by the computer. A plurality of audio files are read from a recording medium, the plurality of audio files are continuously recognized by speech, converted into text data, and the progress of the text data conversion process is displayed for each audio file.

In a sixth aspect based on the fifth aspect, the voice recognition program further comprises: a computer for causing a computer to convert the total file size or total voice data amount of a plurality of voice files to be converted into text data; The data size or audio data amount of the audio file corresponding to the end of the text data conversion and the ratio of the data size or the audio data amount to the total file size or the total audio data amount are calculated, and the calculation result is further displayed. .

[0011] Further, a speech recognition method according to a seventh invention is characterized in that:
Selecting a plurality of audio files from a recording medium on which an audio file having audio data is recorded; continuously converting the selected plurality of audio files into text data; and Displaying for each audio file.

[0012]

Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 1 is a diagram illustrating a configuration of a speech recognition device according to the present embodiment. The voice recognition device according to the present embodiment includes a personal computer (hereinafter, a personal computer) 2. The personal computer 2 is connected via a cable 12 to an audio recording / reproducing apparatus 1 for generating audio data to be subjected to audio recognition and transmitting the generated audio data to the personal computer 2.
Note that the audio recording / reproducing apparatus 1 and the personal computer 2 may perform wireless data communication.

The personal computer 2 has a control unit 4 for controlling each unit.
The control unit 4 is connected to a hard disk (HD) 3 as a recording medium for recording a voice data file to be voice-recognized, a RAM 5, and an operation member 11 such as a mouse as selection means. At the same time, a display device (display means) 10 is connected via a display control device 9. Further, a digital / analog converter (D / A) 8, an amplifier 7, and a speaker 6 are connected as functions for reproducing audio data. The control unit 4 includes a voice recognition unit (voice recognition unit) 4A that can continuously convert voice data in the selected voice file into text data.

FIG. 2 is a diagram showing a display screen displayed on the display device 10 when the speech recognition software stored in the speech recognition section 4A is activated. File list display section 10
3, a plurality of files 104-1 to 104-5 are displayed. Among them, 104-1 is highlighted and indicates that the operator has selected it. After a file to be subjected to voice recognition is selected by the operator, the voice recognition button 102 is operated (clicked) to start voice recognition processing.

FIG. 3 is a main flowchart showing the operation of the control unit in response to an operation by an operator in the first embodiment of the present invention. First, various default values are set in the initial setting in step S1. In the next step S2, it is determined whether or not a file selecting operation has been performed. If YES, the process proceeds to step S5 to highlight the selected file. If the selected file has already been selected, the selection is canceled and the display is returned to the non-reversed display. If the determination in step S2 is NO, or after step S5, step S2
3 to determine whether any of the various buttons displayed in FIG. 2 has been turned on (clicked), and YES
In the case of, the process proceeds to step S4 to execute the command corresponding to the turned on button, and then returns. If the determination in step S3 is NO, the process immediately returns to step S2. In step S3, the voice recognition button 102
Is also detected, and if the voice recognition button 102 is clicked, a voice recognition process is performed in step S4.

FIG. 4 is a flowchart for explaining the details of the voice recognition processing in the first embodiment. Here, it is assumed that a plurality of files are continuously subjected to voice recognition. The voice recognition is enabled only when the high sound quality mode is selected. First, it is determined whether or not a voice file to be recognized has been selected in step S10. If NO, the process proceeds to step S11, for example, "No file has been selected. Select a file and try again. A message such as "." Is displayed in a dialog box.

If the determination in step S10 is YES, the process advances to step S12 to determine whether speech recognition is possible. In the case of NO here, the process proceeds to step S36, where "* .dss (* means an arbitrary file having an extension of dss) among the selected files"
Cannot be recognized because of low sound quality mode. Do you perform voice recognition only for other files? Message is displayed. In the next step S37, it is determined whether or not to continue the processing. If YES, the process proceeds to step S38 to automatically select only a file that can be subjected to voice recognition. If the determination is NO, the process returns.

If the determination in step S12 is YES, or after the process in step S38, the flow advances to step S13 to specify the storage destination (specific folder) of the text file. Here, if the operator does not specify, a default value is set. Next, proceeding to step S14,
A list of files selected when performing voice recognition is displayed together with the processing status.

FIG. 5 is a display screen showing the status of such voice recognition processing. As shown in the figure, since the voice recognition of the selected file has not started yet, all files (Welcome.dss, Welcome2.dss, Welcome3.
dss, Welcome4.dss, Welcome5.dss) are displayed as "Waiting". On the right side of the screen, a confirmation (OK) button 110, a pause button (Pause) 111, and a delete (Delete) button 1
12. A Cancel button 113 is displayed.

After the file list is displayed in step S14, the flow advances to step S15 to start speech recognition processing for the selected file Welcome.dss. Next, the process proceeds to step S15, and the display of the file (Welcome.dss) under voice recognition is changed to "processing".

Next, the process proceeds to step S17 in FIG. 6 to determine whether or not a file selection operation has been performed on the voice recognition processing screen. If YES, the process proceeds to step S18 to highlight the selected file. If the selected file has already been selected, the selection is canceled and the display is returned to the non-reversed display. Step S1
7 is NO, or after the process of step S18, the process proceeds to step S19, where the pause button 11
1, delete button 112, cancel button 113,
It is determined whether any of the confirmation (OK) buttons 110 has been turned on. If, for example, the pause button 111 has been turned on, the process proceeds to step S20 to temporarily stop the voice recognition processing. In this state, the system waits until the pause button is pressed again. When the pause button is pressed again, the determination in step S21 becomes YES, the speech recognition process is restarted in the next step S23, and the process proceeds to step S30.

If the delete button 112 is pressed, the flow advances to step S24 to delete the selected file from the list, and shifts to step S30. Even when a file for which voice recognition is being performed is selected, the process is stopped and then deleted from the list.

If the cancel button 113 is pressed, the flow advances to step S25 to stop the processing of the file currently under voice recognition, delete all files from the list, and end the voice recognition processing.

If the confirmation (OK) button 110 has been pressed, the flow advances to step S26 to minimize the current screen, and then to step S30.

In step S19, a pause button 111, a delete button 112, a cancel button 1
13. If it is determined that none of the confirmation (OK) buttons 110 have been turned on, step S30 is immediately performed.
Move to

While the speech recognition is being performed, the text subjected to the speech recognition processing can be displayed in another window. FIG. 7 shows an example of such a window. While looking at this screen, if the operator determines that the speech recognition rate is not good and the text is not good, the operator presses the delete button 112 to stop the text conversion of the file being processed and perform speech recognition of the next file. It is also possible to shift to processing.

In step S30, it is determined whether or not the voice recognition processing for one file has been completed. If NO, the process returns to step S16. If the determination in step S30 is YES, the process proceeds to step S31, where the extension of the text file is added to the folder specified in step S13. vps
And save.

Then, the flow advances to step S32 to delete the file name for which speech recognition has been performed from the speech recognition processing screen. FIG.
Indicates the screen display at this point. As can be seen from the figure, the processed file (here, Welcom
e.dss, Welcome2.dss) are erased and unprocessed files (Welcome3.dss, Welcome4.ds here)
s, Welcome5.dss) are displayed.

Next, the process moves to the next file selected (step S33), and it is determined whether or not the voice recognition processing for all files has been completed (step S34). If NO, the process returns to step S15 and moves. Perform voice recognition processing of the file. In this way, when the voice recognition processing for all the files is completed, the determination in step S34 becomes YES, the process proceeds to the next step S35, an end display as shown in FIG. 9 is performed, and the process returns.

According to the above-described first embodiment, the file names of a plurality of audio data files and an indication of whether the conversion processing of the audio data files having the file names has not been processed, is being processed, or has been processed. Are notified as a pair, so that the progress of the voice recognition work can be easily grasped even when a plurality of files are voice-recognized at once. Further, by deleting the display of the voice data file for which the voice recognition processing has been completed, it is possible to easily grasp how many unprocessed voice files are present.

Hereinafter, a second embodiment of the present invention will be described.
Here, as in the first embodiment, it is assumed that a plurality of files are continuously subjected to voice recognition.

FIG. 10 shows a second embodiment of the present invention.
5 is a main flowchart showing an operation of a control unit in response to an operation from an operator. Steps S41 to S45 correspond to steps S1 to S5 shown in FIG. 3 and the processing is the same, so that the description here is omitted. In the second embodiment, after selecting a file and highlighting the selected file, it is determined in step S46 whether the file can be recognized by voice. If YES, step S48 is performed.
To make the voice recognition button selectable, and in the case of NO, the process proceeds to the non-selectable step of step S47 to display the voice recognition button in gray tone, for example, so that the selection is invalid even if the operator presses the voice recognition button. Let the operator know that there is something. Note that it is better to perform the processing of steps S36 to S38 in the first embodiment also in step S48. Since the processing in steps S36 to S38 is the same as that in the first embodiment, the description is omitted here. After step S47 or step S48, the process proceeds to step S43.

FIG. 11 is a flowchart for explaining the details of the voice recognition processing in the second embodiment. The file list display in step S51 is the same as that in step S1 in FIG.
4 and the processing before this is the same as that in FIG. 4, and the description thereof is omitted here. After the file list is displayed in step S51, the total file capacity (size) of the plurality of files selected in the next step S52, that is, the total audio data amount is calculated.

The process proceeds to the next step S52 to judge whether or not the calculated total file size is larger than a predetermined size. If YES, the process proceeds to step S54, for example, "Since the total file size is large, It takes about 5 minutes to process. Do you want to continue as it is? " Then, in the next step S55, the process waits for an instruction from the user as to whether or not to continue the process. If the process is to be continued, the process proceeds to step S56; otherwise, the process returns. Steps S53, S54, S5 described above
In No. 5, if the total capacity of the selected file is larger than the predetermined capacity, the voice recognition processing takes a long time, and the intention is to allow the operator to determine whether or not to continue the processing. At this time, the CPU speed,
It is also possible to display the approximate time required for processing from the RAM capacity, the free space of the hard disk (HD) and the total capacity of the selected file.

In step S56, the first file to be subjected to voice recognition is displayed as "processing". Next, step S57
To start speech recognition processing. After the voice recognition is started, the process proceeds to step S58 to notify the operator of the percentage of the entire process that has been completed in step S5.
A ratio (ratio) of the processed file size converted into text data to the total file size (total audio data amount) calculated in step 2, ie, the processed audio data amount is calculated. Proceeding to the next step S59, the display of the bar graph is sequentially corrected based on the ratio obtained in step S58.

FIG. 12 is a display screen showing the status of the voice recognition process. A confirmation (OK) button 210, a pause button 211, a delete button 212, and a cancel button 213 shown in the figure are a confirmation (OK) button 110, a pause button 111, a delete button 112 of the first embodiment,
It has the same function as the cancel button 113. 2
Reference numeral 14 denotes a bar graph display based on a percentage (%), and the tip of the color bar 214A moves to the right as the speech recognition process proceeds.

During the voice recognition process, it is determined whether the voice recognition process for one file has been completed (step S6).
0) The above steps are repeated until the processing of one file is completed. When the processing of one file is completed, the determination in step S60 becomes YES, and the process proceeds to the next step S61 to save a text file as a recognition result. Here, the text file obtained by the speech recognition processing contains. The file is saved in the designated folder with the same name as the audio file with an extension of vps.

In the next step S62, it is determined whether or not the voice recognition for all the files has been completed.
In the case of, the process proceeds to step S64 to switch the display. Here, as shown in FIG. 12, without deleting the processed file, the status display of the file is set to "completed" and the status display of the next file is set to "processing". In the next step S65, the voice recognition processing for the next file is started, and the process returns to step S58 to repeat the above steps. When the voice recognition processing for all the files is completed in this way, the determination in step S62 is YES, the process proceeds to the next step S63, an end display is performed, and the process returns.

In the above-described embodiment, the audio data file recorded on the HD 3 is read and the voice recognition is performed. However, it is also conceivable that the voice recognition processing is performed simultaneously while receiving the voice data from the voice recording / reproducing apparatus. Can be In such a case, by writing information about the total file size in the header of the transmitted audio data frame, the total file size can be known. Therefore, the ratio (ratio) of the processed file size to the total file size should be calculated. Is possible.

According to the above-described second embodiment, the progress of the voice recognition process is indicated by the bar graph display. Work progress can be grasped by comparison with the whole.

[0041]

According to the present invention, it is possible to easily recognize the progress of the voice recognition work even when a plurality of files are voice-recognized at once.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration of a speech recognition device according to an embodiment of the present invention.

FIG. 2 is a diagram illustrating a display screen displayed on a display device when voice recognition software stored in a voice recognition unit is activated.

FIG. 3 is a main flowchart showing an operation of a control unit in response to an operation by an operator in the first embodiment of the present invention.

FIG. 4 is a flowchart illustrating a part of the details of a voice recognition process according to the first embodiment.

FIG. 5 is a display screen showing the status of a voice recognition process.

FIG. 6 is a flowchart for explaining another part of the details of the voice recognition process in the first embodiment.

FIG. 7 is a diagram showing a window displaying text subjected to voice recognition processing.

FIG. 8 is a diagram showing a state where a file name for which speech recognition has been completed is deleted.

FIG. 9 is a diagram showing a display screen when the voice recognition processing is completed.

FIG. 10 is a main flowchart showing an operation of a control unit in response to an operation by an operator in a second embodiment of the present invention.

FIG. 11 is a flowchart illustrating details of a speech recognition process according to the second embodiment.

FIG. 12 is a display screen showing a status of a voice recognition process.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Voice recording / reproducing apparatus, 2 ... Personal computer (PC), 3 ... Recording medium (HD), 4 ... Control part, 5 ... RAM, 6 ... Speaker, 7 ... Amplifier, 8 ... Digital / analog converter (D / A) ), 9: display control device, 10: display device, 11: keyboard, 12: cable.

Claims (7)

[Claims] 1. A recording medium for recording an audio file having audio data, a selection unit capable of selecting a plurality of audio files, and a plurality of audio data of the plurality of audio files selected by the selection unit. A voice recognition unit for converting the text data into text data, and a display unit for displaying the progress of the text data conversion process by the voice recognition unit for each voice file. 2. The method according to claim 1, wherein the display unit determines a file name of the plurality of audio files selected by the selection unit, and determines whether text data conversion processing for the audio file corresponding to the file name has not been processed, is being processed, or has been processed. The speech recognition device according to claim 1, wherein a progress status indicating the following is displayed. 3. The display means displays file names of a plurality of audio files selected by the selection means,
2. The speech recognition apparatus according to claim 1, wherein the display of the file name of the speech file for which the text data conversion processing has been completed is deleted. 4. A method according to claim 1, further comprising: a total audio data amount of the plurality of audio files selected by said selection means, an audio data amount of an audio file corresponding to a text data converted by said voice recognition means, and said total audio data amount. 4. The apparatus according to claim 1, further comprising: a calculating unit configured to calculate a ratio of the audio data amount, wherein the display unit further displays a calculation result of the calculating unit. 5. Voice recognition device. 5. A recording medium recording a processing program for performing a voice recognition process by a computer, the processing program comprising: a plurality of voice data stored on a recording medium storing a voice file having voice data by the computer. A voice recognition processing program for reading a file, continuously performing voice recognition on the plurality of voice files to convert the voice files into text data, and displaying the progress of the text data conversion process for each voice file is recorded. Recording medium. 6. The voice recognition program further causes the computer to transmit a total voice data amount of a plurality of voice files to be converted to text data, a voice data volume of a voice file corresponding to the end of the text data conversion, and The recording medium according to claim 5, wherein a ratio of the audio data amount to the audio data amount is calculated, and the calculation result is further displayed. 7. A step of selecting a plurality of audio files from a recording medium on which an audio file having audio data is recorded; a step of continuously converting the selected plurality of audio files into text data; Displaying the progress of the processing for each audio file.