JP2006251898A - Information processor, information processing method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an information processor for adequately grasping the interest and stance of a conference participant. <P>SOLUTION: A conference history display device (the information processor) 1 comprises: a voice recognition part 3 for generating a writing text based on voice data of a speaker by using voice recognition technique, so as to generate speech data; a speech database 4 for storing the speech data; a speaker setting part 5 for setting the speaker; a tabulating part 6 for tabulating the appearance times of words in the speech data by correspondence to the time axis of the speech data, based on the speech data stored in the speech database 4; and a control part 7 for displaying a graph where the appearance times of the words in the speech data is made to correspond to the time axis of speaker data. Consequently, it is easily recognized who makes a speech, when the speech is made, and what the content of the speech is when subsequently examining the conference. Thus, the interest and stance of each participant is adequately grasped. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an information processing apparatus, an information processing method, and a program.

  2. Description of the Related Art Conventionally, there has been proposed a playback apparatus that can record information such as conferences without omission, and can play back only selected portions by easily specifying information requested by a user later (Patent Document 1).

  Further, Patent Document 2 proposes a speech structure detection and display device that detects and displays the speech structure of participants in a conference. The apparatus described in Patent Document 2 specifies information on a speech section for each speaker, an attitude for each speaker, and who speaks, and detects a flow of speech in a predetermined time section. High reciprocal relations can be displayed.

  Also, a speech structure information presentation device has been proposed for efficiently accessing a desired location when reproducing conference information by identifying and presenting an appropriate speech structure section for each of a plurality of related speech groups. (Patent Document 3).

JP-A-6-343146 JP-A-11-259501 Japanese Patent Laid-Open No. 11-272679

  However, in the prior art, there is a problem that it is difficult to accurately grasp, for example, what and how a certain conference participant talked about at the conference and what interest and stance they have.

  Therefore, the present invention has been made in view of the above problems, and an object thereof is to provide an information processing apparatus, an information processing method, and a program capable of accurately grasping each person's interest and stance.

  In order to solve the above problems, the present invention provides a storage means for storing speech data, and words in the speech data corresponding to the time axis of the speech data based on the speech data stored in the storage means. It is an information processing apparatus provided with the totaling means which totalizes the appearance frequency. According to the present invention, by counting the number of appearances (appearance frequency) of words in the utterance data in correspondence with the time axis of the utterance data, for example, when reviewing the meeting later, who spoke what and when It is easy to understand and can accurately understand the interests and stances of each participant.

  The present invention further includes control means for displaying a graph in which the number of appearances of words in the utterance data is associated with the time axis of the speaker data. Thus, for example, when looking back at a meeting later, it is easy to know who spoke what and when, and it is possible to accurately grasp the interest, attitude, and stance of each participant. The present invention further includes control means for displaying, for each speaker, a graph in which the number of appearances of words in the utterance data is associated with the time axis of the speaker data. The control means displays word names having a high frequency of appearance among words in the utterance data in association with the graph. Thereby, it is possible to know at a glance what word name is spoken.

  The control means displays word names having a high frequency of appearance among words in the utterance data in association with the utterance time in correspondence with the graph. Thereby, it is possible to grasp at a glance what kind of word name is spoken. The control means displays an utterance partner in correspondence with the graph. As a result, it is possible to grasp at a glance who has spoken to. The control means reproduces a corresponding video by performing a predetermined operation on the graph. Thereby, the speaker data can be actually confirmed.

  The present invention further includes setting means for setting speakers to be counted by the counting means. According to the present invention, it is possible to know the number of times a word appears for a specific speaker. The present invention further includes voice processing means for creating the utterance data by generating a transcript from the voice data of the speaker using a predetermined voice recognition technique. By creating a transcript in advance, the results can be displayed quickly. The utterance data is created from voice data detected during a meeting. Thereby, the reflection of the meeting can be supported. The utterance data includes an utterance start time, an utterance end time, a speaker, an utterance content, and an utterance partner.

  The present invention includes a step of acquiring utterance data stored in a predetermined storage means, and a step of counting the number of appearances of words in the utterance data in accordance with the time axis of the utterance data based on the utterance data Is an information processing method. According to the present invention, by counting the number of appearances (appearance frequency) of words in the utterance data in correspondence with the time axis of the utterance data, it is easy to determine who spoke what and when and when to look back on the meeting later. You can understand the interests, attitudes, and stances of each participant. The present invention further includes a step of displaying a graph in which the number of appearances of words in the utterance data is associated with the time axis of the speaker data.

  The present invention includes a step of acquiring utterance data, a step of counting the number of appearances of words in the utterance data based on the utterance data, in correspondence with the time axis, and the number of appearances of words in the utterance data. A program for causing a computer to execute a step of generating information for displaying a graph corresponding to a time axis of speaker data. According to the present invention, by counting the number of appearances (appearance frequency) of words in the utterance data in correspondence with the time axis of the utterance data, it is easy to tell who spoke what and when when looking back at the meeting later. You can understand the interests, attitudes, and stances of each participant.

  According to the present invention, it is possible to provide an information processing apparatus, an information processing method, and a program that can accurately grasp each person's interests and stances.

  Hereinafter, the best mode for carrying out the present invention will be described.

  FIG. 1 is a block diagram of a conference history display device (information processing device) 1. As shown in FIG. 1, the conference history display device 1 includes a conference record database 2, a voice recognition unit 3, an utterance database 4, a speaker setting unit 5, a totaling unit 6, a control unit 7, a display unit 8, and a speaker 9. .

  The conference history display device 1 displays a conference history. The conference record database 2 stores audio data and conference video data of conference participants (speakers) detected during the conference. The voice recognition unit 3 creates utterance data including the utterance contents by generating a transcribed text from the voice data of the speaker stored in the conference recording database 2 using a predetermined voice recognition technology, Utterance data including the utterance completion time, speaker, utterance partner, and utterance content is stored in the utterance database 4. In this way, the utterance data is created from the voice data detected during the conference.

  FIG. 2 shows the contents of the utterance database 4. As shown in FIG. 2, the utterance database 4 includes utterance start time, utterance end time, speaker, utterance content (utterance data), and utterance partner fields. Here, Mr. A, Mr. B, Mr. C, and Mr. D exist as conference participants (speakers). For this reason, there are A, B, C, and D as speaking partners. The utterance content is held as text data. The speaker setting unit 5 is composed of, for example, a keyboard and a mouse. The user can set the speakers to be counted by the counting unit 6 by using the speaker setting unit 5.

  The totaling unit 6 extracts the utterance data spoken by the set speaker from the utterance database 4. FIG. 3 is a diagram showing utterance data extracted from the utterance database 4 when Mr. C is set as the speaker. When C is set as the speaker, the utterance data shown in FIG. 3 is extracted from the utterance database 4 from the utterance database 4. The totaling unit 6 analyzes the text in the “utterance content” field of the extracted utterance data, and totals the number of appearances of words in the utterance data in correspondence with the time axis of the utterance data. Here, “corresponding to the time axis of utterance data” means holding the time of the utterance data or considering the time axis of the utterance data. FIG. 4 shows an example in which the totaling unit 6 extracts only nouns from the text in the utterance content field shown in FIG. As shown in FIG. 4, the totaling unit 6 extracts only nouns in the utterance content field as words.

  The display unit 8 is configured by a display device, for example. Based on the counting result, the control unit 7 displays the total number of words appearing in (for example) a bar graph on the graph and the word having the highest number of appearances on the graph. Is also displayed.

  Next, a display example of the display unit 8 will be described. FIG. 5 is a diagram showing a display example when Mr. C is set as the speaker. As shown in FIG. 5, the control unit 7 displays a bar graph in which the number of appearances of words in the utterance content of the utterance data is associated with the time axis of the speaker data. In FIG. 5, the horizontal axis indicates the conference elapsed time and is displayed so that the conference start time and conference end time can be understood. The vertical axis indicates the number of words spoken by Mr. C. In the window 20, the speaker name (Mr. C) set by the speaker setting unit 5 is displayed.

  In addition, the control unit 7 graphs the word names “cost”, “problem”, “base”, and “xx” that have the highest number of appearances (the number of appearances is higher) among the words in the utterance content of the utterance data. Display in correspondence. Further, the control unit 7 selects the word names having the highest number of appearances among the words in the utterance content of the utterance data as utterance times “15:43:04”, “15:43:45”, “15:44:45”. "Seconds" is displayed in correspondence with the graph. When the control unit 7 displays a word name having a high frequency of appearance among words in the utterance content of the utterance data, the control unit 7 displays an icon B representing Mr. B who is the conversation partner in association with the graph. When the user performs a predetermined operation on the graph such as, for example, moving or clicking the bar 21, the control unit 7 reproduces the corresponding conference video and audio data from the display unit 8 and the speaker 9.

  Next, another display example will be described. FIG. 6 is a diagram illustrating a display example when A, B, and C are set as speakers. As shown in FIG. 6, the control unit 7 displays, for each speaker (A, B, C), a line graph in which the number of appearances of words in the utterance content of the utterance data is associated with the time axis of the speaker data. . In FIG. 6, the horizontal axis indicates the conference elapsed time and is displayed so that the conference start time and conference end time can be understood. The vertical axis indicates the number of words spoken by the speaker. In the window 20, speaker names A, B, and C set by the speaker setting unit 5 are displayed.

  When the control unit 7 displays a line graph in which the number of appearances of words in the utterance content of the utterance data is associated with the time axis of the speaker data, the control unit 7 represents the conversation partners A, B, and C. Icons A, B, and C are displayed in correspondence with the graph. The line L1 of “Mr. A's remark” literally represents “when and how many words Mr. A spoke”. If the face icon appearing on these lines is on the line L1 of “Mr. A's remark”, it represents Mr. A's conversation partner at that time. In the case of a meeting of four people from A to D, a face icon of someone among B, C, and D who is a participant other than Mr. A appears on the line L1 of “Mr. A's remark”. Conversely, Mr. A's face icon does not appear on the line of Mr. A's remarks. L21 is a line representing Mr. B's remarks. There was no remarks at the time before and after that (no remarks for a while after the start of the meeting, but a certain time after that, a little remarks at the end of the meeting) )). When the user performs a predetermined operation on the graph such as, for example, moving or clicking the bar 21, the control unit 7 reproduces the corresponding conference video.

  Next, the operation of the conference history display device 1 will be described. FIG. 7 is an operation flowchart of the conference history display device 1. In step S1, the voice recognition unit 3 creates utterance data by generating a transcribed text from the voice data of the speaker stored in the conference recording database 2 using a predetermined voice recognition technology, and utterance start time, The speaker data is stored in the utterance database 4 corresponding to the utterance completion time, the speaker, and the utterance partner. In step S <b> 2, the user uses the speaker setting unit 5 to set speakers to be aggregated by the aggregation unit 6.

  In step S3, the tabulation unit 6 analyzes the text in the “utterance content” field of the extracted data, and tabulates the number of appearances of words in the utterance content of the utterance data in accordance with the time axis of the utterance data. . In step S4, the control unit 7 displays a graph in which the number of appearances of words in the utterance data is associated with the time axis of the speaker data, as shown in FIGS.

  According to the conference history display device of the present embodiment, when a certain participant is set, data that has been voice-recognized and converted into text is searched in advance, and the result of analyzing the participant's utterance frequency and utterance content is visually displayed. By displaying on the screen, it is easy to know who spoke what and when, and it is possible to accurately and intuitively grasp the interest and stance of each participant. That is, when looking back at a meeting where at least the voice of the participant is recorded, it is possible to easily and intuitively grasp the interest and attitude of each participant.

  The information processing apparatus according to the present invention is realized using, for example, a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like. Each step of the information processing method is realized by installing the program from a hard disk drive, a portable storage medium such as a CD-ROM, DVD or flexible disk, or downloading from a communication circuit, and the CPU executing this program Is done. The program further includes a step of acquiring utterance data, a step of counting the number of appearances of words in the utterance data in correspondence with the time axis based on the utterance data, and the number of appearances of words in the utterance data. A computer is caused to generate the information for displaying the graph corresponding to the time axis of the speaker data. The speech database 4 corresponds to storage means for storing speech data, and the speech recognition unit 3 corresponds to speech processing means.

  Although the preferred embodiments of the present invention have been described in detail above, the present invention is not limited to the specific embodiments, and various modifications, within the scope of the gist of the present invention described in the claims, It can be changed.

It is a block diagram of a meeting history display device. It is a figure which shows the content of the speech database. It is a figure which shows the speech data extracted from the speech database 4 when C is set as a speaker. This is an example in which only nouns are extracted from text. It is an example of a display when Mr. C is set as a speaker. It is a display example when A, B, and C are set as speakers. It is an operation | movement flowchart of a meeting history display apparatus.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Conference history display apparatus 2 Conference record database 3 Voice recognition part 4 Speech database 5 Speaker setting part 6 Total part 7 Control part 8 Display part 9 Speaker

Claims (14)

Storage means for storing utterance data;
An information processing apparatus comprising: an aggregation unit that aggregates the number of appearances of words in the utterance data in correspondence with a time axis of the utterance data based on the utterance data stored in the storage unit. The information processing apparatus according to claim 1, further comprising a control unit that displays a graph in which the number of appearances of words in the utterance data is associated with a time axis of the speaker data. The information processing apparatus according to claim 1, further comprising a control unit that displays, for each speaker, a graph in which the number of appearances of words in the utterance data is associated with the time axis of the speaker data. The information processing apparatus according to claim 2, wherein the control unit displays word names having a large number of appearances in the words in the utterance data in association with the graph. 4. The information processing apparatus according to claim 2, wherein the control unit displays a word name having a large number of appearances among words in the utterance data in association with the utterance time in association with the graph. . The information processing apparatus according to claim 2, wherein the control unit displays an utterance partner in association with the graph. The information processing apparatus according to claim 2, wherein the control unit reproduces a corresponding video by performing a predetermined operation on the graph. The information processing apparatus according to claim 1, further comprising setting means for setting a speaker to be aggregated by the aggregation means. 2. The information processing apparatus according to claim 1, further comprising voice processing means for creating the utterance data by generating a transcript from the voice data of a speaker using a predetermined voice recognition technology. The information processing apparatus according to claim 1, wherein the utterance data is created from voice data detected during a meeting. The information processing apparatus according to claim 1, wherein the utterance data includes an utterance start time, an utterance end time, a speaker, an utterance content, and an utterance partner. Obtaining utterance data stored in a predetermined storage means;
An information processing method comprising: counting the number of appearances of words in the utterance data based on the utterance data in correspondence with a time axis of the utterance data. The information processing method according to claim 12, further comprising a step of displaying a graph in which the number of appearances of words in the utterance data is associated with a time axis of the speaker data. Obtaining utterance data;
Based on the utterance data, counting the number of appearances of words in the utterance data in correspondence with the time axis;
The program for making a computer perform the step which produces | generates the information for displaying the graph which made the frequency | count of appearance of the word in the said speech data correspond to the time-axis of the said speaker data.

Images