JP6865701B2 - Speech recognition error correction support device and its program - Google Patents

Description

The present invention relates to a voice recognition error correction support device that supports voice recognition error correction and a program thereof.

When using the audio of audio material (including video and audio material) recorded in program coverage as characters, the work of transcribing the audio into characters is indispensable. Usually, this work is performed by listening to the voice of the material recorded by the worker and inputting characters with the keyboard of the terminal or the like. At this time, the worker frequently repeats playing and stopping the sound, and re-listens to the sound at the same place many times. It is said that it takes about 6 times as long as the work time.

Conventionally, as a technology to support the voice transcription work, voice recognition processing is performed for each sentence (cell) in which the input voice is divided into arbitrary units, and the recognition result obtained by the voice recognition processing and the corresponding voice A technique for correcting an error in speech recognition processing is disclosed (see Patent Document 1).
In this technique, the voice is reproduced in cell units in the voice recognition process, the operator corrects the recognition result in cell units, and the cells are corrected by operating a general text editor. In addition, with this technology, the operator does not need to remember any special operation, and after modifying the cell, the voice is played from the beginning of the cell to check whether the operator has corrected the recognition result correctly. It was.

Further, as a technique for supporting the conventional speech transcription work, a technique for associating speech recognition results for each word and correcting the speech for each word is disclosed (see Patent Documents 2 and 3).
This technique is effective for error correction such as subtitle broadcasting that requires real-time performance and for correcting recognition results with few errors.

Japanese Unexamined Patent Publication No. 2015-184564 Japanese Unexamined Patent Publication No. 2004-226910 Japanese Unexamined Patent Publication No. 2005-228178

Since the technique disclosed in Patent Document 1 reproduces the sound and corrects the recognition result on a cell-by-cell basis, it is necessary to confirm whether or not the sound of the corrected part matches the corrected result even if the corrected part is small. In addition, it is necessary to play the sound from the beginning of the cell.
Therefore, this technique has a problem that a waiting time is generated until the sound of the correction target portion in the middle of the cell is reproduced. Further, in this technology, since it is necessary for the operator to distinguish the voice corresponding to the recognition result in the cell, there is a problem that it becomes difficult to associate the voice with the correction target when the recognition result becomes poor. It was.

Further, in the technique of correcting the speech recognition result on a word-by-word basis, such as the technique disclosed in Patent Documents 2 and 3, it is possible to quickly correct the recognition result and confirm the speech. However, if there is a recognition error across multiple words, it is necessary to specify the words in order and make corrections, which complicates the procedure and takes time to get used to the operation. ..

Therefore, the present invention is a voice recognition error correction support device and a program thereof, which can quickly reproduce the voice of the correction target portion and correct the voice recognition error with a simple operation when correcting the voice recognition error. The challenge is to provide.

To solve the above problems, the speech recognition error correction support device according to the present invention, there is provided a speech recognition error correction support device for correcting errors in speech recognition for voice content, the recognition result dividing means, recognition The configuration includes a result display control means, an error correction means, and a voice reproduction means.

In such a configuration, the voice recognition error correction support device uses the recognition result dividing means to segment the recognition result based on a predetermined standard based on the recognition result of the voice as text data and the time information for each word constituting the recognition result. It is divided into.

Then, the speech recognition error correction support device by the recognition result display control means display the Rubo Tan specify whether to display a word sequence contained in the segment with item information. Further, the speech recognition error correction support device, by the recognition result display control means performs the selection of buttons, or to deploy the word sequence of segments to display the editing area, whether the control and hide the editing area .. As a result, the recognition result display control means does not display all the voice recognition results, but causes the operator to specify the segment to be edited by the item list, and expands the word string of the target segment into the editing area. And present it to the operator.

Then, the speech recognition error correction support device by the error correction means corrects the error of the segment in the editing area. In this case, the error correcting means, is reproduced by the audio reproducing means audio Turkey content to correspond to the time information from by word position specified in the editing area. As a result, the error correction means reproduces the voice from the word at the designated position so that the operator can quickly confirm the recognition result or the voice corresponding to the correction result.
The voice recognition error correction support device can be operated by a voice recognition error correction support program for operating the computer as each of the above-mentioned means.

The present invention has the following excellent effects.
According to the present invention, since the voice recognition result of the material content is divided and a list of items is displayed, it is possible to quickly select the recognition result for which the voice recognition error is to be confirmed by a simple operation.
Further, according to the present invention, since the corresponding voice is reproduced by a simple operation of designating the position of the word in the editing area, it is possible to quickly find an error in the voice recognition result and confirm the correction.
Thereby, the present invention can correct the error of the speech recognition result without requiring a special skill.

It is a block block diagram which shows the structure of the voice recognition error correction support device which concerns on embodiment of this invention. It is explanatory drawing for demonstrating the memorized content memorized by the material information storage means. It is a screen block diagram which shows an example of the material content selection screen which selects a material content. It is a screen block diagram which shows an example of the item list screen which shows the list of the item which divided the voice recognition result of the material content. It is a screen block diagram which shows the example which expanded the voice recognition result in the edit area on the item list screen. It is explanatory drawing for demonstrating an example of an editing work in an editing area. It is explanatory drawing for demonstrating the example which changes the display attribute of a word of an edit area in conjunction with voice reproduction. It is explanatory drawing for demonstrating the example which presents the operation content of the editing work in an editing area. It is explanatory drawing for demonstrating the example which specifies the repeated reproduction of audio in an editing area. It is a flowchart which shows the segment information generation operation which generates the voice recognition result of the voice recognition error correction support apparatus which concerns on embodiment of this invention in segment units. It is a flowchart which shows the segment information presenting operation which presents the voice recognition result of the voice recognition error correction support device which concerns on embodiment of this invention to a display device in segment units. It is a flowchart which shows the segment correction operation which corrects a recognition result while performing the voice reproduction of the voice recognition error correction support device which concerns on embodiment of this invention. It is a block block diagram which shows the structure of the voice recognition error correction support device which concerns on embodiment of the modification of this invention.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[Configuration of voice recognition error correction support device]
First, the configuration of the voice recognition error correction support device 1 according to the embodiment of the present invention will be described with reference to FIG.
The voice recognition error correction support device 1 supports correction of voice recognition errors in at least material content including voice. In the present embodiment, the material content is a content composed of video and audio, for example, a broadcasting material.

As shown in FIG. 1, the voice recognition error correction support device 1 includes a material content input means 10, a voice recognition means 11, a recognition result dividing means 12, an item information extraction means 13, a material information storage means 14, and the like. The editing means 15 and the transcription result output means 16 are provided.

The material content input means 10 inputs material content.
The material content input means 10 may, for example, input material content from an external storage medium, or may input material content via a communication line.
The material content input means 10 outputs the voice of the input material content to the voice recognition means 11. Further, the material content input means 10 writes and stores the input material content (video / audio) in the material information storage means 14 in order to use it for the correction work in the editing means 15 described later.

After writing the material content in the material information storage means 14, the material content input means 10 notifies the voice recognition means 11 of the completion of writing the material content, and the voice recognition means 11 sends the material information storage means 14 to the material information storage means 14. You may read the voice.

The voice recognition means 11 recognizes the voice of the material content input by the material content input means 10 and generates a recognition result which is text data and time information for each word constituting the recognition result.
The voice recognition means 11 performs voice recognition by using a language model, an acoustic model, and a pronunciation dictionary (not shown), and generates a recognized word and time information indicating an elapsed time from the beginning of the voice of the word. The voice recognition means 11 outputs the generated recognition result word and time information to the recognition result dividing means 12. As the voice recognition method in the voice recognition means 11, for example, a method of recognizing a word string from the voice disclosed in Japanese Patent Application Laid-Open No. 2010-175765 and outputting the result may be used.

The recognition result dividing means 12 divides the recognition result (word string) recognized by the voice recognition means 11 according to a predetermined standard. Hereinafter, each block of the division recognition results generated by the recognition result division means 12 is referred to as a segment.
Any standard can be predetermined as the division standard used by the recognition result dividing means 12.
For example, a silent section of voice can be used as a reference for division. In this case, the recognition result dividing means 12 detects a silent section from the voice stored in the material information storage means 14 by power or the like, which is an acoustic feature amount, and transmits the recognition result by the voice recognition means 11 before and after the silent section. To divide.

Further, for example, a cut point of an image can be used as a reference for division. In this case, the recognition result dividing means 12 detects as a cut point a frame in which the image features of the adjacent frames are significantly different from the predetermined reference from the video stored in the material information storage means 14, and the time of the cut point is reached. The recognition result is divided before and after.

Further, for example, meta information added in advance to the material content may be used as a reference for division. The meta information includes GPS (Global Positioning System) position information (geotag) and the like. In this case, the recognition result dividing means 12 divides the recognition result at a time when the place where the material content is photographed or the sound is collected differs depending on the position information.

The recognition result dividing means 12 outputs the segment obtained by dividing the voice recognition result to the item information extracting means 13. Further, the recognition result dividing means 12 writes and stores a word and its time information in the material information storage means 14 for each segment.

The item information extracting means 13 extracts the feature words included in the segment as an item for each segment divided by the recognition result dividing means 12.
This characteristic word is a characteristic word contained in the segment. For example, the item information extracting means 13 extracts words that characterize a segment by the TF-IDF method (TF: Term Frequency, word appearance frequency, IDF: Inverse Document Frequency, reverse document frequency). TF-IDF is a kind of weight related to a word in a document (segment in this embodiment), and is mainly used in fields such as information retrieval and sentence summarization.
Specifically, the item information extraction means 13 calculates the appearance frequency tf (w, s) of the word w in the segment s by the following equation (1).

In this formula _{(1), n w, s} is the number of occurrences of in segment s of a word _w, Σ t∈s n _{t, s} represents the sum of all the number of occurrences of a word in the segment s.
Further, the item information extraction means 13 calculates the inverse document frequency idf (w) of a certain word w by the following equation (2).

In this formula (2), N indicates the total number of segments in the material content, and df (w) indicates the number of segments of the material content in which a certain word w appears (total number of segments [total number of documents]).
Then, the item information extraction means 13 has the largest product of the tf value of the equation (1) and the idf value of the equation (2) for each word in the segment, as shown in the following equation (3). Alternatively, a word larger than a predetermined reference value is set as a characteristic word of the segment.

The item information extraction means 13 writes and stores the extracted items in the material information storage means 14 in association with the segments.
The item information extracting means 13 may perform morphological analysis of the segment and extract nouns and proper nouns as feature words without using the TF-IDF method.

Further, when the material content includes an image, the item information extracting means 13 may extract a thumbnail image from the image of the time interval corresponding to the segment in addition to the feature word. For example, the item information extracting means 13 extracts the first frame of the video in the time interval corresponding to the segment as a thumbnail image. The item information extraction means 13 writes and stores the extracted thumbnail image in the material information storage means 14 in association with the segment.

The material information storage means (storage means) 14 stores material content to be corrected for an error in voice recognition and various types of information obtained by dividing the material content into segments. The material information storage means 14 can be composed of a general storage medium such as a hard disk or a semiconductor memory.

Here, the material information stored in the material information storage means 14 will be specifically described with reference to FIG. 2 (see FIG. 1 as appropriate).
As shown in FIG. 2, the material information storage means 14 stores material contents (video / audio) A, B ... To be corrected for a voice recognition error. The material contents (video / audio) A, B ... Are stored by the material content input means 10.

Further, as shown in FIG. 2, the material information storage means 14 stores information obtained by dividing the voice recognition result into segments for each material content.
In the example of FIG. 2, segments (identification information a1, a2, ..., b1, ...) Are associated with each identification information (here, file names A, B, ...) Of the material content.
Each segment contains a plurality of words w and time information t, and each segment is associated with each other.
The word w and the time information t for each segment are information obtained by dividing the word and time information associated with the voice recognition means 11 by the recognition result dividing means 12.

In addition, each segment includes an item k and a thumbnail image g. Item k is a feature word extracted by the item information extracting means 13. The thumbnail image g is a frame image extracted from the video of the material content by the item information extracting means 13 corresponding to the time information at the beginning of the segment.
Here, the material content and the segment obtained by dividing the voice recognition result of the material content are stored in the same storage means, but may be stored in different storage means.
Returning to FIG. 1, the configuration of the voice recognition error correction support device 1 will be continued.

The editing means 15 is for the operator to correct the voice recognition result stored in the material information storage means 14 by using the correction terminals (input device 2, display device 3, speaker 4) connected to the outside. is there. The display device 3 of the correction terminal may be configured to include a touch panel.
As shown in FIG. 1, the editing means 15 includes a material content selection means 150, a recognition result display control means 151, an error correction means 152, and a video / audio reproduction means 153.

The material content selection means 150 selects the material content to be modified. For example, as shown in FIG. 3, the material content selection means 150 includes a material content selection button 301 for selecting any of the material contents A, B, and C stored in the material information storage means 14. The screen 30 is displayed on the display device 3. Then, the material content selection means 150 selects the material content to be corrected by pressing the selection button 301 on the material content selection screen 30. The material content selection means 150 outputs identification information such as a file name of the selected material content to the recognition result display control means 151.

The recognition result display control means 151 displays an item and a selection button for specifying whether or not to display the word string included in the segment for each segment, and displays the word string of the segment by pressing the selection button. It controls whether or not.

Here, a screen example displayed by the recognition result display control means 151 will be described together with the control contents with reference to FIGS. 4 and 5 (see FIG. 1 as appropriate).
As shown in FIG. 4, the recognition result display control means 151 displays the item list screen 31 on the screen of the display device 3.
The item list screen 31 is composed of a selection button 311, an item display field 312, a thumbnail image display area 313, a timetable display field 314, and a scroll bar display field 315.

The selection button 311 is a button for selecting whether or not to display a word string for each segment.
The item display field 312 is an area for displaying the items extracted in the segment. The recognition result display control means 151 reads out the item (item k in FIG. 2) corresponding to the segment from the material information storage means 14, and displays it in the item display column 312.
The thumbnail image display area 313 is an area for displaying the thumbnail image extracted in the segment. The recognition result display control means 151 reads out a thumbnail image (thumbnail image g in FIG. 2) corresponding to the segment from the material information storage means 14, and displays it in the thumbnail image display area 313.

The timetable display column 314 is a column for displaying a timetable indicating the segment position on the time axis of the material content. The recognition result display control means 151 generates and displays a time table with reference to the time information (time information t in FIG. 2) of the segment of the material information storage means 14.
The scroll bar display field 315 is a field for displaying a scroll bar indicating which part of the segment is displayed when the item list does not fit on the screen. The recognition result display control means 151 updates the item list on the screen by moving the scroll bar up and down.
By displaying the item list screen 31 in this way, the operator can confirm the items, and can easily select the segment for which the voice recognition result is to be confirmed, as compared with the case where the voice recognition result is displayed at once. can do.

On this item list screen 31, recognition result display control is performed by clicking the mouse of the input device 2 performed by the operator or pressing the selection button (“open” in FIG. 4) 311 by touching the touch panel of the display device 3. The means 151 displays an editing area 316 (see FIG. 5) for correcting a word string of a segment on the item list screen 31.

FIG. 5 is a screen example showing the item list screen 31B displaying the edit area 316.
In the item list screen 31B, the moving image display area 313B and the editing area 316 are displayed in the selected segment with respect to the item list screen 31 described with reference to FIG.

The moving image display area 313B is an area for reproducing the material content corresponding to the segment. The recognition result display control means 151 refers to the time information (time information t in FIG. 2) of the segment of the material information storage means 14 at the timing when the segment is selected, and sets the first frame of the video of the corresponding material content. It is displayed in the moving image display area 313B. By clicking the image area of the moving image display area 313B with a mouse or the like or pressing the reproduction start button st, the recognition result display control means 151 instructs the video / audio reproduction means 153 to reproduce the material content.

The editing area 316 is an area to be edited by displaying the word string corresponding to the segment. The recognition result display control means 151 develops a word string (row of the word w in FIG. 2) corresponding to the segment stored in the material information storage means 14 in the editing area 316.
At this time, the recognition result display control means 151 uses the selection button 311 as a button (“close” in FIG. 4) for hiding the editing area 316. Then, by pressing the selection button (“close” in FIG. 4) 311 the recognition result display control means 151 hides the editing area 316, sets the moving image display area 313B as the thumbnail image display area 313, and sets the item in FIG. The display is returned to the list screen 31.
Returning to FIG. 1, the configuration of the voice recognition error correction support device 1 will be continued.

The error correction means 152 corrects an error in the word string of the segment in the editing area 316 (FIG. 5) by the editing operation of the operator. The error correction means 152 functions as a general text editor (screen editor) in the editing operation for correcting the word string. However, the error correction means 152 has a function of reproducing a voice when correcting a word string.

Specifically, the error correction means 152 reproduces the voice from the selected word in the editing area 316 (FIG. 5) by clicking the mouse or touching the touch panel. Further, during the sound reproduction, the error correction means 152 stops the sound reproduction by selecting an arbitrary position again.

FIG. 6 is an explanatory diagram for explaining an example of editing work in the editing area.
For example, when "March" is selected in the editing area 316 of FIG. 6, the error correction means 152 refers to the time information of the segment of the material information storage means 14 (time information t of FIG. 2) and responds. The video / audio reproduction means 153 is instructed to reproduce the audio from the position of the material segment to be used. At this time, in conjunction with the audio, the video corresponding to the audio reproduction time may be reproduced in the moving image display area 313B (FIG. 5).
Here, when the operator finds an error (here, "grouper cold") and clicks the corrected portion with the mouse, the error correcting means 152 stops the voice reproduction and displays the cursor C. Then, the error correction means 152 corrects the error "grouper cold" as "chilly" by the editing operation of the operator. Then, the error correction means 152 replaces the erroneous word stored in the material information storage means 14 with the corrected word. As a result, the voice recognition error correction support device 1 can omit the save operation after correction by the operator.

Further, the error correction means 152 reproduces the sound from the word position designated by a mouse click or the like.
FIG. 7 is an explanatory diagram for explaining an example of changing the display attribute of a word in the editing area in conjunction with voice reproduction. For example, as shown in FIG. 7, when a portion in the editing area 316 where the sound reproduction is to be started is selected with a mouse or the like, the error correction means 152 uses the time information of the segment of the material information storage means 14 (FIG. 2). With reference to the time information t), the video / audio reproduction means 153 is instructed to reproduce the audio from the selected word until the instruction to stop the reproduction is given.
Then, as shown in FIG. 7, the error correction means 152 indicates each word corresponding to the reproduced voice in conjunction with the reproduction of the voice so as to clearly indicate the reproduction position of the voice and the presence / absence of reproduction in the segment. Change the display attribute of the display part. For example, the error correction means 152 displays the word corresponding to the voice in black-and-white inversion or in a predetermined color.

At this time, the error correction means 152 presents feedback information of the operation performed by the operator on the screen. For example, as shown in FIG. 8, the error correction means 152 displays a pop-up message pop1 indicating the start of voice reproduction at the selected word position, and pop-up message pop2 indicating the end of voice reproduction at the word position where the voice has stopped. Is displayed. As a result, even if the operator is unfamiliar, he / she can grasp his / her own operation contents and can perform the operation with peace of mind.

Further, the error correction means 152 can repeatedly reproduce the voice corresponding to the designated word or word string.
For example, as shown in FIG. 9, in the editing area 316, by selecting a word or a word string for which voice is to be reproduced with a mouse or the like (black and white inverted area in the figure), the error correction means 152 displays a pop-up menu pm. Then, by selecting "Repeat playback", the sound of the corresponding word or word string is repeatedly played.
Returning to FIG. 1, the configuration of the voice recognition error correction support device 1 will be continued.

The video / audio reproduction means 153 reproduces the video and audio of the material content. The video / audio reproduction means 153 reproduces the material content (video / audio) from the position designated by the recognition result display control means 151 or the error correction means 152.

The transcription result output means 16 outputs the voice recognition result (transcription result) corrected by the editing means 15 to the outside.
The transcription result output means 16 specifies the file name of the material content or the identification number of the segment in the material content, and the word of the corresponding material content or segment stored in the material information storage means 14. Read the column and output it.

By configuring the voice recognition error correction support device 1 as described above, the voice recognition error correction support device 1 can confirm the recognition result word and the original voice by a simple text editing operation. , Speech recognition errors can be corrected. In addition, the voice recognition error correction support device 1 can partially correct errors in the material content on a segment-by-segment basis.
The voice recognition error correction support device 1 can be operated by a voice recognition error correction support program for operating the computer as each of the above-mentioned means.

[Operation of voice recognition error correction support device]
Next, the operation of the voice recognition error correction support device 1 according to the embodiment of the present invention will be described with reference to FIGS. 10 to 12. Here, as the operation of the voice recognition error correction support device 1, the segment information generation operation of generating the recognition result by voice recognition for the material content in segment units and the recognition result are presented to the display device 3 in segment units. The segment information presentation operation and the segment correction operation for correcting the recognition result while performing voice reproduction will be described.

(Segment information generation operation)
First, the segment information generation operation of the voice recognition error correction support device 1 will be described with reference to FIG. 10 (see FIG. 1 as appropriate).
In step S1, the material content input means 10 inputs the material content for voice recognition. At this time, the material content input means 10 writes and stores the input material content in the material information storage means 14.

In step S2, the voice recognition means 11 recognizes the voice of the material content input in step S1 and generates the recognition result which is text data and the time information for each word constituting the recognition result in association with each other.

In step S3, the recognition result dividing means 12 divides the recognition result recognized in step S2 into segments according to a predetermined reference, for example, a video cut point, a silent section of audio, or the like. At this time, the recognition result dividing means 12 writes and stores the material content in the material information storage means 14 in association with the word of the recognition result and the time information in segment units.

In step S4, the item information extracting means 13 extracts the feature word included in the segment as an item for each segment divided in step S3, and extracts the thumbnail image from the video corresponding to the segment. At this time, the item information extraction means 13 writes and stores the extracted items and thumbnail images in the material information storage means 14 in association with the segments.
As a result of the above operation, the voice recognition error correction support device 1 stores the material content and various information obtained by dividing the material content into segments in the material information storage means 14, as shown in FIG.

(Segment information presentation operation)
Next, the segment information presentation operation of the voice recognition error correction support device 1 will be described with reference to FIG. 11 (see FIG. 1 as appropriate).
In step S10, the material content selection means 150 displays the material content selection screen 30 (see FIG. 3) including a selection button for selecting any of the material contents stored in the material information storage means 14. Display in.

In step S11, the material content selection means 150 waits until the selection button is pressed on the screen (No in step S11), and when the selection button is pressed (Yes in step S11), controls after step S12. Control is transferred to the recognition result display control means 151 to be performed.

In step S12, the recognition result display control means 151 specifies whether to display an item and a word string included in the segment for each segment based on various information stored in the material information storage means 14. The item list screen 31 (see FIG. 4) including the selection button to be selected is displayed on the display device 3.

In step S13, the recognition result display control means 151 waits until the selection button (open) is pressed on the item list screen (No in step S13).
On the other hand, when the selection button (open) is pressed (Yes in step S13), in step S14, the recognition result display control means 151 sets the editing area 316 corresponding to the selected segment as shown in FIG. The word string which is the recognition result corresponding to the segment which is displayed and stored in the material information storage means 14 is expanded in the editing area 316.

After this operation, the voice recognition error correction support device 1 shifts to a state in which the operator can correct the editing result on the screen. When the edit area 316 is displayed by pressing the selection button (open), the recognition result display control means 151 can hide the edit area 316 by pressing the select button (close) at an arbitrary timing. However, the illustration of this non-display operation is omitted. Further, the description of the reproduction operation of the material content in the moving image display area 313B of the item list screen 31B (see FIG. 5) will be omitted here.
By the above operation, the voice recognition error correction support device 1 can correct the voice recognition error in the material content in segment units.

(Segment correction operation)
Next, the segment correction operation of the voice recognition error correction support device 1 will be described with reference to FIG. 12 (see FIG. 1 as appropriate). Since the segment correction operation is an arbitrary procedure performed by the operator, an example of an operation in which the voice reproduction and the correction operation are performed together will be described here.

In step S20, the error correction means 152 selects a word or a word string for which voice is to be reproduced in the editing area 316 (FIG. 5) by clicking the mouse of the operator or touching the touch panel. At this time, the error correction means 152 refers to the time information of the segment of the material information storage means 14 via the video / audio reproduction means 153, and reproduces the sound of the time corresponding to the word or the word string. This allows the operator to compare and confirm the voice and the voice-recognized word string.

In step S21, the error correction means 152 accepts the designation of the position of the correction portion by clicking the mouse of the operator or touching the touch panel. At this time, the error correction means 152 stops the reproduction of the sound if the sound has not been reproduced to the end of the word string, or if the sound is being repeatedly reproduced and the sound is being reproduced.

In step S22, the error correction means 152 displays a cursor at a designated position in the editing area, and corrects the recognition error by the operator's editing work such as character deletion and character insertion. Here, the error correction means 152 updates the word of the material information storage means 14 with the correction result.

In step S23, the error correction means 152 accepts the designation of the position of the corrected portion by clicking the mouse of the operator or touching the touch panel. At this time, the error correction means 152 refers to the time information of the segment of the material information storage means 14 via the video / audio reproduction means 153, and reproduces the sound of the time corresponding to the word or the word string. As a result, the operator can confirm whether or not the correction result is correct.

Although not shown, if the operator has confirmed in step S23 that the corrected portion has not yet been corrected correctly, the process returns to step S21 and the operation is repeated.
By the above operation, the voice recognition error correction support device 1 can quickly reproduce the voice of the correction target portion when correcting the voice recognition error, and can correct the voice recognition error with a simple operation.

Although the embodiment of the present invention has been described above, the present invention is not limited to this embodiment.
Here, the material content has been described as including video and audio, but the material content may be audio only.
In that case, the item information extracting means 13 may extract only the items and not the thumbnail images. Further, the video / audio reproduction means 153 may be an audio reproduction means for reproducing only the audio.

Further, here, the correction terminal (input device 2, display device 3, speaker 4) is directly connected to the voice recognition error correction support device 1, but these are in the form of being connected via a network. It doesn't matter.

Further, the voice recognition error correction support device 1 may be configured to include a plurality of correction terminals. In that case, the recognition result display control means 151 exclusively controls the segment being modified by one modification terminal so that another modification terminal does not select it as a modification target. For example, in another modification terminal, a selection button is pressed. Hide it.

Further, the editing means 15 of the voice recognition error correction support device 1 provides a user interface for performing screen control as a server for correcting the recognition result, and a plurality of correction terminals connected via a network use the user interface. It may be made to function as a client that operates through. This makes it possible to correct voice recognition errors at multiple points via the network.

Further, the voice recognition error correction support device 1 may be provided with the voice recognition means 11 externally.
For example, the voice recognition error correction support device 1B shown in FIG. 13 may be configured. The voice recognition error correction support device 1B is externally provided with the voice recognition means 11 of the voice recognition error correction support device 1 (FIG. 1) as a voice recognition device. In this case, the recognition result dividing means 12 inputs the voice recognition result output from the voice recognition means 11 and the time information for each word constituting the recognition result via the recognition result input means 17 which is an input interface. do it.
The voice recognition error correction support device 1B can also be operated by the voice recognition error correction support program for operating the computer as each of the above-mentioned means.

1,1B Voice recognition error correction support device 10 Material content input means 11 Voice recognition means 12 Recognition result division means 13 Item information extraction means 14 Material information storage means (memory means)
15 Editing means 150 Material content selection means 151 Recognition result display control means 152 Error correction means 153 Video / audio reproduction means (audio reproduction means)
16 Transcription result output means 17 Recognition result input means

Claims (9)

It is a voice recognition error correction support device that corrects voice recognition errors in the voice contained in the content.
A recognition result dividing means for dividing the recognition result into segments according to a predetermined standard based on the recognition result of the voice which is text data and the time information for each word constituting the recognition result.
A button for specifying whether or not to display the word string included in the segment is displayed together with the item information, and by selecting the button, the edit area is displayed and the word string of the segment is expanded, or the edit area is displayed. A recognition result display control means that controls whether to hide or not,
An error correction means for correcting an error in the segment in the editing area,
A sound reproduction means for reproducing the sound corresponding to the segment of the editing area is provided.
The recognition result dividing means divides the recognition result at the change point of the position information or the time information included in the content.
The error correction means is a voice recognition error correction support device, characterized in that the voice of the content corresponding to the time information from the word position designated in the editing area is reproduced by the voice reproduction means. It is a voice recognition error correction support device that corrects voice recognition errors in the voice contained in the content.
A recognition result dividing means for dividing the recognition result into segments according to a predetermined standard based on the recognition result of the voice which is text data and the time information for each word constituting the recognition result.
A button for specifying whether or not to display the word string included in the segment is displayed together with the item information, and by selecting the button, the edit area is displayed and the word string of the segment is expanded, or the edit area is displayed. A recognition result display control means that controls whether to hide or not,
An error correction means for correcting an error in the segment in the editing area,
A sound reproduction means for reproducing the sound corresponding to the segment of the editing area is provided.
The content includes an image, and the recognition result dividing means divides the recognition result at a cut point of the image.
The error correction means is a voice recognition error correction support device, characterized in that the voice of the content corresponding to the time information from the word position designated in the editing area is reproduced by the voice reproduction means. It is a voice recognition error correction support device that corrects voice recognition errors in the voice contained in the content.
A recognition result dividing means for dividing the recognition result into segments according to a predetermined standard based on the recognition result of the voice which is text data and the time information for each word constituting the recognition result.
A button for specifying whether or not to display the word string included in the segment is displayed together with the item information, and by selecting the button, the edit area is displayed and the word string of the segment is expanded, or the edit area is displayed. A recognition result display control means that controls whether to hide or not,
An error correction means for correcting an error in the segment in the editing area,
An audio reproduction means for reproducing the audio corresponding to the segment in the editing area, and
Each of the segments is provided with an item information extraction means for extracting a feature word as the item information by the TF-IDF method from the words included in the plurality of the segments.
The recognition result display control means displays a list of the item information including a button for specifying whether or not to display the word string included in the segment.
The error correction means is a voice recognition error correction support device, characterized in that the voice of the content corresponding to the time information from the word position designated in the editing area is reproduced by the voice reproduction means. It is a voice recognition error correction support device that corrects voice recognition errors in the voice contained in the content.
A recognition result dividing means for dividing the recognition result into segments according to a predetermined standard based on the recognition result of the voice which is text data and the time information for each word constituting the recognition result.
A button for specifying whether or not to display the word string included in the segment is displayed together with the item information, and by selecting the button, the edit area is displayed and the word string of the segment is expanded, or the edit area is displayed. A recognition result display control means that controls whether to hide or not,
An error correction means for correcting an error in the segment in the editing area,
A sound reproduction means for reproducing the sound corresponding to the segment of the editing area is provided.
The error correction means reproduces the sound of the content corresponding to the time information from the word position designated in the editing area by the sound reproduction means , and any word in the editing area during the sound reproduction of the content. By designating the position, the playback of the voice in the voice reproduction means is stopped, a pop-up message indicating the start of voice playback is displayed at the word position specified in the editing area, and the voice is displayed at the word position where the voice is stopped. speech recognition error correction support device according to claim you to view the pop-up message indicating the playback end. Claims 1 to claim 1, wherein the error correction means stops the reproduction of the voice in the voice reproduction means by designating an arbitrary word position in the editing area during the voice reproduction of the content. The voice recognition error correction support device according to any one of 3. Any one of claims 1 to 5 , wherein the error correction means changes the display attribute of a word in the editing area corresponding to the reproduced voice in conjunction with the voice reproduction of the content. The voice recognition error correction support device described in the section. The claim is characterized in that the error correction means repeatedly reproduces the voice of the content corresponding to the time information of the word designated in the editing area or the word string of the designated section by the voice reproducing means. The voice recognition error correction support device according to any one of claims 1 to 6. It is a voice recognition error correction support device that corrects voice recognition errors in the voice contained in the content.
A recognition result dividing means for dividing the recognition result into segments for each change in the utterance content based on the recognition result of the voice which is text data and the time information for each word constituting the recognition result.
A button for specifying whether or not to display the word string included in the segment is displayed together with the item information, and by selecting the button, the edit area is displayed and the word string of the segment is expanded, or the edit area is displayed. A recognition result display control means that controls whether to hide or not,
An error correction means for correcting an error in the segment in the editing area,
A sound reproduction means for reproducing the sound corresponding to the segment of the editing area is provided.
The error correction means is a voice recognition error correction support device, characterized in that the voice of the content corresponding to the time information from the word position designated in the editing area is reproduced by the voice reproduction means. A voice recognition error correction support program for causing a computer to function as the voice recognition error correction support device according to any one of claims 1 to 8.

Images