JP6524242B2 - Speech recognition result display device, speech recognition result display method, speech recognition result display program - Google Patents

Description

  Embodiments of the present invention relate to a voice recognition result display device, a voice recognition result display method, and a voice recognition result display program.

  2. Description of the Related Art Conventionally, there has been provided a technology for performing voice recognition of voice data input from a client (smart phone, PC, etc.) terminal, converting it into text, and displaying it as a recognition result on a display screen of the client terminal. For example, techniques for displaying speech in a meeting or the like in time series with a time axis, or displaying a speech section and a silent section separately are also provided.

Patent No. 5685702

  The problem to be solved by the present invention relates to a voice which can confirm the theme of the conversation of each time zone, the agenda, etc. by a simple method on the basis of automatically extracted keywords regarding a series of utterances performed in the past. It is providing a recognition result display device.

  The speech recognition result display device according to the embodiment includes a keyword extraction unit for extracting a character string included in speech text data at a predetermined timing from speech text data which is a result of speech recognition processing on speech data, speech text data, keyword Based on the storage unit for recording the character string extracted by the extraction means and the display request of the speech recognition result from the client terminal, the storage unit is searched for the character string extracted from the voice text data in the predetermined time specified from the client terminal A keyword search unit is provided, and the character string searched by the keyword search unit is displayed on the client terminal screen as a keyword.

BRIEF DESCRIPTION OF THE DRAWINGS The block diagram which shows the whole structure of the speech recognition result display system which concerns on 1st Embodiment. The figure which shows an example of the screen display of the user terminal which concerns on 1st Embodiment. FIG. 2 is a view showing an example of voice data and related information according to the first embodiment. FIG. 5 is a view showing an example of keywords extracted from a speech recognition result according to the first embodiment. The figure which shows an example of the keyword display displayed on the screen of the user terminal which concerns on 1st Embodiment. FIG. 5 is a view showing an example of keywords extracted from a speech recognition result according to the first embodiment. The figure which shows an example of the keyword display displayed on the screen of the user terminal which concerns on 1st Embodiment. FIG. 6 is a view showing an example of keywords to be excluded from display among keywords extracted from a speech recognition result according to the first embodiment. The figure which shows an example of the keyword display displayed on the screen of the user terminal which concerns on 1st Embodiment. The figure which shows the processing flow of the speech recognition result display system which concerns on 1st Embodiment. The figure which shows an example of the keyword display displayed on the screen of the user terminal which concerns on 1st Embodiment. The figure which shows an example of the speech recognition result displayed on the screen of the user terminal which concerns on 1st Embodiment. The figure which shows an example of the speech recognition result displayed on the screen of the user terminal which concerns on 1st Embodiment. The figure which shows an example of the speech recognition result displayed on the screen of the user terminal which concerns on 1st Embodiment. The figure which shows the flow of the audio | voice text data display process of the voice recognition result which concerns on 1st Embodiment. The figure which shows an example of the speech recognition result displayed on the screen of the user terminal which concerns on 1st Embodiment.

First Embodiment
Hereinafter, a first embodiment of the present invention will be described with reference to the drawings.

  FIG. 1 is a block diagram showing the overall configuration of a speech recognition result display system according to the first embodiment. As shown in FIG. 1, the voice recognition display system includes a user terminal (client terminal) 100, a voice recognition result display device 200, and a voice recognition server 300, which are connected via a network 400, respectively. The voice recognition server 300 includes an input unit 301, a voice recognition processing unit 302, and an output unit 303, receives voice data input from the user terminal 100 via the voice recognition result display device 200, and the voice recognition processing unit 302 The processed speech recognition result (speech text data) is transmitted to the speech recognition result display device 200.

  The user terminal 100 is, for example, a smartphone, a tablet, a PC or the like, and includes an input / output unit 101, a display control unit 102, and a display unit 103. The input / output unit 101 sends to the voice recognition result display device 200 a request for sending voice from a microphone (not shown), a request for sending voice recognition results to be displayed on the display unit 103, and switching the voice recognition result displayed on the screen based on the request Send. The display unit 103 displays the recognition result of the voice transmitted from the input / output unit 101 to the voice recognition result display device 200.

  The voice data transmitted from the input / output unit 101 to the voice recognition result display device 200 is sent together with identification information for specifying a speaker and a speech time. The transmission request information of the speech recognition result to be displayed on the display unit 103 transmitted from the input / output unit 101 includes the date and time indicating the display request range. The date and time indicating the display request range is acquired based on the user (client) operation on the screen displayed when the user terminal 100 as shown in FIG. 2 is activated. For example, it is possible to display a time axis starting from the time of activation based on a clock built in the user terminal 100, and to obtain the date and time indicating the display request range based on the user operation on the displayed time axis.

  FIG. 2 is an example of a screen displayed when the user terminal 100 is activated. In the display example shown in FIG. 2, the time axis 501 displayed on the screen 500 is from 13 o'clock to 15 o'clock on July 31, 2015, and the time designated by the user by operating the pointer 502 is 14 It has become time. The user terminal 100 specifies a display request range to be requested to the voice recognition result display device 200 from the time axis 501 displayed on the screen 500 at the time of operation by the user operating the pointer 502 specifying 14 o'clock. Time information for specifying the required range is transmitted from the input / output unit 101 to the speech recognition result device 200. In the example of FIG. 2, the time information for specifying the display range is from 13:00 to 15:00 on July 31, 2015. The display range of the time axis 501 shown in FIG. 2 can be switched to a predetermined range such as one day, half a day, one hour, etc. by an operation on the user terminal 100. In FIG. 2, the operation for the display range in the user terminal 100 is an example of two hours.

  The display control unit 102 controls to display on the display unit 103 the speech recognition result requested to be transmitted to the speech recognition result display device 200 via the input / output unit 101. In the present embodiment, the display content and the display position are adjusted based on the keyword received as the display target information from the voice recognition result display device 200 and the information indicating the display position, and the display unit 103 displays the adjusted display content and the display position.

  The voice recognition result display device 200 includes a voice data input unit 201, a voice data output unit 202, a keyword extraction unit 203, a keyword importance degree calculation unit 204, a storage unit 205, and a keyword search unit 206.

  The voice data input unit 201 receives voice data transmitted from the input / output unit 101 of the user terminal 100 and registers the voice data in the storage unit 205 and performs voice recognition on the voice data for the voice recognition server 300 to perform voice recognition processing. Send to server 300. In addition, the result (voice text data) processed by the voice recognition processing unit 302 of the voice recognition server 300 is received and registered in the storage unit 205.

  FIG. 3 is a diagram showing an example of registration of voice data registered in the storage unit 205. As shown in FIG. As shown in FIG. 3, voice data input from the user terminal 100 is registered together with identification information for specifying a speaker and speech time. Furthermore, the result of the recognition process performed by the speech recognition server 300 is registered in association.

  The voice data output unit 202 searches and obtains voice data registered in the storage unit 205 via the voice data input unit 201 in response to a request from the user terminal 100, and transmits the voice data to the user terminal 100. The user terminal 100 can reproduce voice by receiving voice data.

  The keyword extraction unit 203 extracts a character string included in the speech recognition result registered in the storage unit 205, and specifies a character string to be a keyword to be displayed on the user terminal 100. The part of speech of the character string to be extracted may be a part of speech that can be a keyword, for example, a noun or a verb.

  FIG. 4 is a diagram showing the result of keyword extraction. Extraction of a character string to be a keyword is executed by the keyword extraction unit 203 at preset time intervals (once a day, once a half day, once a hour, etc.). The execution result of the keyword extraction is recorded in the storage unit 205 together with the extracted time with the extracted character string as a keyword as shown in FIG. FIG. 4 shows an example in which a character string as a keyword extracted between 13 o'clock and 15 o'clock on July 31, 2015 is recorded. As shown in FIG. 4, it indicates that the character string “Tokyo” is extracted at 13: 5 on July 31, 2015. The string for homework is extracted at 14:40 on July 31, 2015. The string for setting is extracted at 13:15 on July 31, 2015. The string for education is extracted at 14:00 on July 31, 2015. It shows that it was done. For example, when the character string "Tokyo" is extracted a plurality of times by one execution operation of the keyword extracting unit 203, the number of appearances is counted instead of recording each time, and the keyword importance calculating unit 205 described later reflects the weight. Do.

  In addition, the search range for keyword extraction varies depending on the relationship with the execution timing of the keyword extraction unit described above, such as one day, half a day, and one hour for the search target.

  Here, a method has been described in which the keyword extraction unit 203 automatically extracts a character string from the recognition result of voice data registered in the storage unit 205 to specify a keyword, but the user registers a specific keyword in advance. When the extraction unit 203 is executed, it may be searched whether or not the keyword registered by the user is included, and the number of appearances of the keyword may be counted as described above.

  The keyword importance degree calculation unit 204 calculates a weight based on the appearance frequency of the character string extracted by the keyword extraction unit 203, and sets the calculated result as a weight to the extracted character string as shown in FIG. This weighting is not limited to the number of occurrences, but, for example, in the processing of the keyword extraction unit 203, a plurality of the same character strings are included in one utterance of voice data for a predetermined time, or an utterance of one user If the same character string is included in the utterances of a plurality of people, or if the same character string is included in all of the voice data for a predetermined time, the appearance location in the utterance may also be taken into consideration. Good. By weighting based on the appearance frequency, it becomes possible to extract important keywords from the extracted character string. As in the case of the keyword extraction, the user may set weighting.

  The storage unit 205 includes voice data input from the voice data input unit 201, text data of voice recognition results processed by the voice recognition server 300, keywords extracted by the keyword extracting unit 203, and a keyword importance degree calculating unit 204. Record the calculated weights.

  The keyword search unit 206 searches for the keyword extracted by the keyword extraction unit 203 in response to the display request of the speech recognition result from the input / output unit 101 of the user terminal 100, and the display unit 103 of the user terminal 100 Display keywords. When the keyword search unit 206 receives the date and time indicating the display request range from the input / output unit 101 of the user terminal 100, the keyword search unit 206 refers to the keyword extraction result shown in FIG. The time at which the keyword is extracted is compared, and the keyword included in the display request range transmitted from the user terminal 100 is specified. For example, the case of receiving from 13:00 to 15:00 on July 31, 2015 as the display request range from the input / output unit 101 of the user terminal 100 will be described as an example. The keyword search unit 206 refers to the time when the keyword extracted as the keyword extraction result shown in FIG. 4 is extracted, and “Tokyo”, “homework”, “homework” as “keywords included from 13 o'clock to 15 o'clock on July 31, 2015. Identify "setting", "education". Then, as a result of the display request from the user terminal 100, the keyword, the keyword extraction time, and the weight are transmitted to the user terminal 100. The input / output unit 101 of the user terminal 100 receives the information, and the display control unit 102 causes the display unit 103 to display the information.

  Subsequently, a specific display method of the speech recognition result displayed on the display unit 103 of the user terminal 100 will be described.

  FIG. 5 is a view showing a display example of a result searched by the keyword search unit 206. As shown in FIG. In FIG. 5, the above keyword search unit 206 searches from 13:00 to 15:00 on July 31, 2015 as the search range, and displays the extracted four keywords (Tokyo, homework, setting, education) . As shown in FIG. 4, each keyword is displayed by changing the display size and the display format based on the weight value set for the keyword. As shown in FIG. 4, the weight of the four keywords is 0.95 for "Tokyo", 1.0 for "Homework", 0.3 for "Setting", and 0.32 for "Education". Here, the weight uses a numerical value between 0 and 1, indicating that the closer the value to 1, the more important the keyword, and change the form with the other keyword with the highest weight value "homework" It is displayed in the shape of a star. The other three keywords are all displayed in a circular shape, but the importance is changed by changing the size of the circle according to the weight value. The shape of the display is an example, and a display method such as changing a color can be used as long as the display can specify a keyword having a high degree of importance among a plurality of keywords. There is no limitation on the display method.

  Further, the display position of each keyword is determined based on the time when the keyword shown in FIG. 4 is extracted. In the example of FIG. 4, “Tokyo” is 13: 5 on July 31, 2015, and is displayed near 13: 5 as shown in FIG. 5. Similarly, “Homework” is around 14:40 on July 31, 2015, “Setting” is around 13:15 on July 31, 2015, “Education” is on July 14, 2015. Displayed near the hour. From this display, it is understood that there are many utterances including homework as a keyword and many utterances including homework as a keyword around 14:40 in the utterance of 13:00 to 15:00 on July 31, 2015 . In this way, by changing the display format according to the degree of importance and displaying the keywords along the time axis, it is possible to easily see what keywords have been included in each time zone. become able to.

  In the present embodiment, the keyword extraction unit 203 of the voice recognition result display device 200 automatically extracts the keyword from the voice data, and the extracted keyword is displayed on the user terminal 100. However, the number of keywords is large Since the number of keywords that can be displayed on the display screen of the user terminal 100 is limited, there is a possibility that keywords with low weights are not displayed on the screen. Therefore, the user may be able to make settings for excluding from display targets keywords that the user does not need. The method will be briefly described with reference to FIGS.

  FIG. 6 is a diagram showing the result of the keyword extraction, which is the same as the result of the keyword extraction as in FIG. 4 although the registered keywords are different. Although five keywords extracted as shown in FIG. 6 are registered, four keywords can be displayed as shown in FIG. 7 from the relationship of display areas of the display unit 103 of the user terminal 100. Therefore, "setting" which is the keyword with the lowest weight is not displayed. Here, by setting a keyword to be excluded from the display target as shown in FIG. 8, even if the weight is high, if it is registered as a negative keyword, it is excluded from the display target and the weight can not be displayed instead. Display high keywords. FIG. 9 is an example in which a play is registered as a keyword to be excluded, and based on this exclusion setting, “setting” is displayed instead of “play”.

  Here, the processing flow of the speech recognition result display system according to the present embodiment will be described with reference to FIG. FIG. 10 is a diagram showing a processing flow of the speech recognition result display system according to the first embodiment.

  The user terminal 100 transmits the user's speech (speech data) from a microphone or the like from the input / output unit 101 to the speech recognition result display device 200 together with identification information for specifying a speaker and speech time (step S1). The voice recognition result display device 200 transmits the voice data received by the voice data input unit 201 to the voice recognition server 300, and stores the voice data, identification information for identifying the speaker, and the speech time in the storage unit 205 (step S2) ). The speech recognition server 300 causes the speech recognition processing unit 302 to execute recognition processing of the speech data received by the input unit 301, and transmits the text data of the recognition result to the speech recognition result display device 200 (step S3). The voice recognition result display device 200 records the voice text data of the recognition result received from the voice recognition server 300 in association with the voice data recorded in step S1 (step S4). The voice recognition result display device 200 searches for a character string included in voice text data recorded in the storage unit 205 at time intervals preset by the keyword extraction unit 203. The character string extracted as the search result is recorded in the storage unit 205 together with the time when the character string is extracted. When the character string already extracted and recorded in the process of searching for the character string is searched, the number of appearances is counted and recorded (step S5). The keyword importance calculator 204 of the voice recognition result display device 200 sets weights for each character string based on the appearance frequency of the character string for the character string extracted in step S5 (step S6).

  The user terminal 100 transmits a keyword search request including date and time information indicating a display request range from the input / output unit 101 to the speech recognition result display device 200 based on the keyword display request operation of the speech recognition result by the user (step S7) . Based on the date and time indicating the display request range of the keyword search request received from the user terminal 100, the voice recognition result display device 200 extracts the character string (keyword) recorded in the storage unit 205 in the process of step S5. In comparison, keywords extracted within the range of request date and time are identified. Then, the specified keyword, the extraction time of the keyword, and the weight set in the process of step S6 are transmitted to the user terminal 100 (step S8). The user terminal 100 adjusts the display position from the keyword extraction time by the display control unit 102 and adjusts the display size from the weight by using the display control unit 102 with the keywords received from the voice recognition result display device 200 by the input / output unit 101 It displays on 203 (step S9).

  Next, display switching and other display methods for the voice recognition result displayed on the display unit 103 of the user terminal 100 based on the processing described in the flowchart of FIG. 10 will be described with reference to FIGS.

  FIG. 11 is a diagram showing an example of keyword display of speech recognition results. FIG. 12 is a view showing an example of a display to which a speech recognition result corresponding to a keyword is added by a user operation on the display of FIG.

  FIG. 11 shows the result (keyword) of the search based on the request from the user terminal 100 searched by the keyword search unit 206 of the speech recognition result display device 200. Specifically, it is the display screen 600 of the user terminal 100 on which the keyword included in the utterance at 10 o'clock to 12 o'clock on July 31, 2015 is displayed. As a keyword, a business trip is displayed on Friday, and a pointer 602 operated by the user indicates 11 o'clock. In the display screen 600 shown in FIG. 11, the position on the time axis of the pointer 602 is transmitted from the user terminal 100 to the voice recognition result display device 200 as needed by the user's operation of moving the pointer 602 on the time axis 601. The speech recognition result apparatus 200 refers to the speech time recorded in the storage unit 205 together with the speech data shown in FIG. 3 and performs the speech when there is speech for the time indicated by the pointer received from the user terminal 100. The user information and the utterance content are transmitted to the user terminal 100 and displayed.

  A display example of the search result is shown in FIG.

  As shown in FIG. 12, the display control unit 102 controls based on the information of the search result received by the user terminal 100 from the voice recognition result display device 200, and the display unit 103 includes the utterance content 603 including the keyword (Friday) Is displayed at a position corresponding to the time of utterance. As the uttered content, the utterer, the time when the uttered, and the uttered content are displayed. The display content is an example, and a summary may be displayed instead of the uttered content, or other information may be displayed. In the above, the time is selected by the pointer, and the information of the utterance performed at the time is displayed, but the user selects the user terminal 100 by selecting it by touching the keyword on the screen. Information of a keyword can be transmitted to the speech recognition result display device 200. Then, the speech recognition result display device 200 searches for speech including the keyword from the speech recognition result (speech text data) recorded in the storage unit 205 together with the speech data shown in FIG. Information on the utterance may be displayed in the form of a balloon or the like on the time axis of the screen of the user terminal 100 by transmitting information including the utterance time related to the utterance.

  Here, only one utterance is displayed, but when a plurality of utterances are searched as a result of the search, all utterance contents are displayed. The keywords not selected by the user operation may be displayed in gray, or the selected keywords may be displayed highlighted, or the selected keywords and the non-selected keywords may be displayed separately. .

  Subsequently, another display example will be described with reference to FIGS. 13 and 14. FIG. 13 and FIG. 14 show an example of displaying the amount of speech in the time zone displayed on the screen together with the keyword display of the speech recognition result.

  In FIGS. 13 and 14, the keyword search unit 206 of the speech recognition result display device 200 displays the search result (keyword) based on the request from the user terminal 100. FIG. Specifically, it is the display screen 600 of the user terminal 100 on which the keyword included in the utterance at 13:00 to 15:00 on July 31, 2015 is displayed. Although the display example which displays only the keyword is described in FIG. 5, the number of utterances included in the range of the time axis (601) displayed on the user terminal 100 is changed in triangle height in FIG. The time zone is visualized (speech number information 604). Further, FIG. 14 expresses the number of utterances in each time zone by using a line graph (the number of utterances information 605) instead of the triangle display. Regarding the number of utterances, in the processing of the voice recognition result display device 200 in response to the keyword search request from the user terminal 100, the date and time indicating the display request range received from the user terminal 100 in parallel with the above keyword search by the keyword search unit 206 Then, based on the utterance time recorded in the storage unit 205 together with the audio data shown in FIG. 3, the number of records of the audio data included in the display request range is counted, and the predetermined time unit (for example, 30 minutes unit) The number of records is transmitted to the user terminal 100. The user terminal 100 displays the result adjusted by the display control unit 102 on the display unit 103 based on the information received from the voice recognition result display device 200.

  Here, with reference to FIG. 15, the processing flow of the audio | voice text display of the speech recognition result which concerns on this embodiment is demonstrated. FIG. 15 is a diagram showing a processing flow of speech recognition result display according to the first embodiment.

  The user terminal 100 transmits the designated keyword and the date and time corresponding to the display position of the designated keyword to the speech recognition result device 200 in accordance with the user's designation operation for the keyword acquired from the speech recognition result display device 200 displayed on the display unit 103. (Step S11). The voice recognition result display device 200 uses voice text data corresponding to voice data uttered at the date and time recorded in the storage unit 205 based on the designated keyword and date and time received from the user terminal 100 by the keyword search unit 206. The voice text data including the designated keyword is searched (step S12). The voice text data including the designated keyword is transmitted to the user terminal 100 together with the accompanying information such as the speaker and the speech time (step S13). The user terminal 100 adjusts the display position as data including the designated keyword and the voice text data received from the voice recognition result display device 200 by the display control unit 102, and displays it in the vicinity of the designated keyword displayed on the display unit 103. Do.

  Next, another display example will be described with reference to FIG. FIG. 16 is an example of a keyword search screen 700 for searching for a target keyword from past keywords.

  In the example of FIG. 16, an input area 701 of a keyword to be searched is provided at the top of the screen, and a display area 702 including a time axis 702a in units of months as a vertical axis and a time axis 702b in units of time as a horizontal axis. By vertically scrolling the screen, it is possible to select a month to be displayed. In addition, it is possible to switch 12 hours display / 24 hours display by pinch in / pinch out of the screen, and display time zone (0 to 12 o'clock / 12 to 24 o'clock) by scrolling horizontally during 12 hours display. Can be switched.

  When a keyword is input in the keyword input area 701 at the top of the screen and the magnifying glass icon is touched, a display request for the voice recognition result including the input keyword and the month / time indicating the selected display request range is the voice recognition result display device Sent to 200. Based on the received display request of the speech recognition result, the keyword search unit 206 searches the keywords extracted by the keyword extraction unit 203 for the corresponding keyword within the display request range, and transmits the search result to the user terminal 100. . The input / output unit 101 of the user terminal 100 receives the search result, and the display control unit 102 displays a screen as shown in FIG. 16 on the display unit 103.

  In the example of FIG. 16, a circle mark is displayed at the location of the month / time when the input keyword (“business trip” in the example of FIG. 16) is uttered. The size of the circle indicates the number of utterances including the searched keyword. The larger the number of utterances, the larger the circle is displayed, and the smaller the number of utterances, the smaller circle is displayed. Further, when the circle mark displayed in the display area 702 is touched, a keyword list of that time zone as shown in FIG. 2 can be displayed.

  As described above, in the present embodiment, the character recognition result can be displayed as a character string included in the utterance for each time zone as a keyword, and it is easy to determine what kind of theme or topic uttered in which time zone. Can be grasped.

  Further, in the present embodiment, by weighting according to the number (appearance frequency) of the keywords included in the utterance, the size and shape of the display of the keywords can be changed and displayed, and the utterance related to the displayed keywords is It is possible to easily grasp the concentrated time zone.

  Further, in the present embodiment, since the utterance content including the keyword displayed by the operation on the screen can be displayed, it is possible to easily grasp the specific utterance content.

  Further, in the present embodiment, the speech recognition result can be added to the display of the keywords included in the utterance for each time zone, and the number of utterances for each time zone can be displayed, and the time zone actively discussed in the meeting etc. Can be grasped.

  Further, in the present embodiment, since the user can set keywords to be displayed, only necessary keywords can be displayed even when the number of keywords that can be displayed is restricted due to the relationship with the display area. .

  In the present embodiment, the display control of the speech recognition result has been described as being executed by the user terminal 100, but may be performed by the speech recognition result display device 200.

  Note that the method described in the above embodiment can be executed as a program that can be executed by a computer, such as a magnetic disk (floppy (registered trademark) disk, hard disk, etc.), optical disk (CD-ROM, DVD, etc.), magneto-optical disk (CD, etc.) It can also be stored and distributed in a storage medium such as MO) or semiconductor memory.

  Here, as the storage medium, any storage format may be used as long as it can store a program and can be read by a computer.

  In addition, an operating system (OS) operating on a computer based on an instruction of a program installed in the computer from a storage medium, MW (middleware) such as database management software, network software, etc. realize this embodiment. A part of each process may be executed.

  Furthermore, the storage medium in the present embodiment is not limited to a medium independent of the computer, but also includes a storage medium obtained by downloading and storing or temporarily storing a program transmitted by a LAN, the Internet, or the like.

  Further, the storage medium is not limited to one, and the processing in the present embodiment may be executed from a plurality of media, and the storage medium in the present embodiment may have any configuration.

  Note that the computer in the present embodiment executes each process in the present embodiment based on a program stored in a storage medium, and a single device such as a personal computer and a plurality of devices are connected to a network. It may be any configuration such as a system.

  Further, each storage device of the present embodiment may be realized by one storage device or may be realized by a plurality of storage devices.

  The computer in the present embodiment is not limited to a personal computer, but also includes an arithmetic processing unit, a microcomputer, etc. included in an information processing device, and generically refers to devices and devices capable of realizing the functions of the present embodiment by a program. ing.

  While the embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope and the gist of the invention, and are included in the invention described in the claims and the equivalent scope thereof.

100 user terminal 101 input / output unit 102 display control unit 103 display unit 200 voice recognition result display device 201 voice data input unit 202 voice data output unit 203 keyword extraction unit 204 keyword importance degree calculation unit 205 ... storage unit 206 ... keyword search unit 300 ... speech recognition server 301 ... input unit 302 ... speech recognition processing unit 303 ... output unit 400 ... networks 500, 600, 700 ... speech recognition result display screen 501, 601 ... time axis 502, 602 ... Operation pointer 603 ... Speech information 604, 605 ... Speech number information

Claims (7)

A speech recognition result display device for displaying a result of speech recognition processing for speech data on a client terminal screen, comprising:
Keyword extraction means for extracting a character string included in the voice text data at predetermined timing from the voice text data which is the result of the voice recognition process;
A storage unit that records the voice text data, the character string, and a time when the character string is extracted by the keyword extraction unit;
Based on the display request of the voice recognition result from the client terminal, the predetermined time designated from the client terminal is compared with the time when the character string is extracted, and the time when the character string is extracted is the designated predetermined time And keyword search means for searching the storage unit for the character string included in
A voice recognition result display device characterized by displaying the character string searched by the keyword search means as a keyword on the client terminal screen. The apparatus has a keyword importance degree calculation unit that calculates a weight for the character string based on the appearance frequency of the character string extracted by the keyword extraction unit.
The storage unit associates the weight calculated by the keyword importance degree calculation means with the character string and records the weight.
Based on the display request of the voice recognition result from said client terminal, wherein the character string searched by the keyword searching unit for the specified predetermined time from the client terminal, by changing the format in accordance with the weight client The speech recognition result display device according to claim 1, wherein the speech recognition result is displayed on a terminal screen.   3. The voice recognition result display device according to claim 2, wherein a format according to the weight displayed on the client terminal screen is a size or a color. It said storage unit records the time obtained by extracting the character string from the audio-text data in the keyword extraction means,
Based on said display request of the speech recognition result from the client terminal, corresponding to the time of extracting the character string the string searched by the keyword searching unit for the specified predetermined time from the client terminal as a keyword The voice recognition result display device according to any one of claims 1 to 3, wherein the voice recognition result is displayed at a position.   The keyword search means searches the storage unit for voice text data including a keyword designated from the client terminal based on a display request of speech information of speech recognition result by the keyword designation operation displayed on the client terminal screen. 5. The voice recognition result display device according to claim 1, wherein the voice text data is displayed in the vicinity of a designated keyword on the client terminal screen. A voice recognition result display method performed by a voice recognition result device for displaying a result of voice recognition processing for voice data on a client terminal screen, comprising:
Extracting a character string included in the voice text data at predetermined timing from the voice text data which is the result of the voice recognition process;
Recording in the storage unit the voice text data, the character string extracted from the voice text data, and a time when the character string is extracted in the extracting step;
Based on the display request of the voice recognition result from the client terminal, the predetermined time designated from the client terminal is compared with the time when the character string is extracted, and the time when the character string is extracted is the designated predetermined time Retrieving the character string included in the storage unit from the storage unit;
A voice recognition result display method comprising displaying a character string retrieved from the storage unit on the client terminal screen as a keyword. A program executed by a speech recognition result display device for displaying the result of speech recognition processing for speech data on a client terminal screen, comprising:
A keyword extraction function of extracting a character string included in the voice text data at predetermined timing from the voice text data which is the result of the voice recognition process;
A recording function for recording the voice text data, the character string, and a time when the character string is extracted by the keyword extraction unit;
Based on the display request of the voice recognition result from the client terminal, a predetermined time designated by the client terminal is compared with the time when the character string is extracted from the information recorded by the recording function, and the character string is A keyword search function of searching for the character string whose extracted time is included in the designated predetermined time ;
A voice recognition result display program, wherein the character string searched by the keyword search function is displayed on the client terminal screen as a keyword.