JP6626029B2 - Information processing apparatus, information processing method and program - Google Patents

Description

  An embodiment of the present invention relates to an information processing device, an information processing method, and a program.

  In recent years, with the globalization of culture and economy, opportunities for communication between people having different languages as their mother tongues have increased. Therefore, a system has been proposed in which, even when a speech recognition technology and a machine translation technology are combined and a conference is held using different native languages, the minutes are created in a certain language.

Japanese Patent No. 4466666

  However, in the related art, the speech recognition result itself is created as minutes, or the minutes are created by directly translating the speech recognition result. Therefore, the minutes are redundant, and it is difficult for the user to quickly grasp necessary information. In addition, since the result of machine translation is not always correct, a manual check is required to ensure accuracy.

  It is an object of the present invention to provide an information processing apparatus, an information processing method, and a program that can smoothly create minutes (summary) in multiple languages.

  An information processing apparatus according to an embodiment includes a search unit, a display control unit, and a generation unit. The search unit is configured to output a first sentence described in a first language and a second sentence described in a second language among one or more text data described in one or more languages, obtained by recognizing a voice. Is searched for text data that conforms to the summary template associated with, and the template in which the matching text data is retrieved is output. The display control unit displays the output template in association with the matching text data. The generation unit generates an abstract based on the template.

FIG. 1 is a block diagram illustrating a configuration example of the information processing apparatus according to the first embodiment. FIG. 2 is a diagram illustrating an example of the summary template. FIG. 3 is a diagram illustrating an example of the summary template. FIG. 4 is a diagram illustrating an example of a structure of data stored in the recognition result storage unit. FIG. 5 is a diagram illustrating an example of the translation result. FIG. 6 is a diagram illustrating an example of the generation screen. FIG. 7 is a diagram illustrating an example of a generation screen that displays a plurality of candidates in a selectable manner. FIG. 8 is a diagram illustrating an example of the generation screen after changing the value of the variable. FIG. 9 is a diagram illustrating an example of the generated minutes. FIG. 10 is a flowchart illustrating an example of the voice recognition process according to the first embodiment. FIG. 11 is a flowchart illustrating an example of minutes candidate generation processing according to the first embodiment. FIG. 12 is a flowchart illustrating an example of the minutes generation process according to the first embodiment. FIG. 13 is a block diagram illustrating a configuration example of an information processing apparatus according to the second embodiment. FIG. 14 is a diagram illustrating an example of the editing operation. FIG. 15 is a flowchart illustrating an example of the minutes editing process according to the second embodiment. FIG. 16 is an explanatory diagram illustrating an example of a hardware configuration of the information processing apparatus according to the first or second embodiment.

  Hereinafter, preferred embodiments of an information processing apparatus according to the present invention will be described in detail with reference to the accompanying drawings.

  In the following, a case will be described as an example where speech input to a conference system (conference device) is recognized and minutes of the conference are generated from the recognition result. Applicable systems are not limited to conference systems. For example, the present invention may be applied to a system (apparatus) that generates information that summarizes a recognition result from a speech recognition result or the like.

  Further, in the following, description will be given by taking an example in which the languages used by the meeting participants are Japanese and English, and the language used by the minutes creator is Japanese. The language to be processed is not limited to these two languages, and various languages can be targeted. It is assumed that the user inputs these pieces of linguistic information into the system in advance at the time of participating in a meeting or preparing minutes.

  In the following embodiments, it is assumed that communication means between languages during a conference is separately prepared. For example, the interpreter may separately participate in the meeting and perform the interpreting, or may perform the translation via a voice translating device that translates and outputs the input voice.

(1st Embodiment)
The information processing apparatus according to the first embodiment searches for a template that matches a speech recognition result from templates of minutes that correspond to a plurality of sentences described in different languages, and uses the searched template to Generate a record. The searched template is displayed on a display unit or the like, and is modified as necessary. With such a configuration, it is possible to easily create the contents necessary and sufficient for creating the minutes. In addition, high-precision translation using templates in which sentences in a plurality of languages are associated makes it possible to simultaneously provide minutes in multiple languages.

  FIG. 1 is a block diagram illustrating a configuration example of the information processing apparatus 100 according to the first embodiment. As shown in FIG. 1, the information processing device 100 is connected to a terminal device 200 via a network or the like. The network may be either a wired network or a wireless network, or a network in which both are mixed. The network is, for example, the Internet.

  The information processing device 100 may be physically realized as one device (a server device, a personal computer, or the like), or may be realized as a logical device including a plurality of physical devices. The information processing apparatus 100 is realized by, for example, a virtual computer constructed on a cloud environment. The terminal device 200 is realized, for example, as a device using a cloud environment such as a personal computer and a mobile terminal.

  The terminal device 200 is mainly used for displaying information for generating minutes in response to an instruction from the information processing device 100 and transmitting information for generating minutes to the information processing device 100. The processing executed by the terminal device 200 may be configured to be executed in the information processing device 100. For example, a display unit may be provided in the information processing apparatus 100, and information for generating minutes may be displayed on the display unit. In this case, there is no need to provide the terminal device 200.

  The terminal device 200 includes a display unit 221, a communication control unit 201, and a display control unit 202.

  The display unit 221 is a device that displays various information such as images. The display unit 221 can be realized by a liquid crystal display, a touch panel, or the like. The display content on the display unit 221 is controlled by the display control unit 202.

  The communication control unit 201 controls communication with an external device such as the information processing device 100. For example, the communication control unit 201 receives, from the information processing apparatus 100, information for displaying a generation screen used for generating minutes. The communication control unit 201 transmits information specified by the user to the information processing device 100.

  The display control unit 202 controls the display of various types of information on the display unit 221 according to an instruction from a display control unit 105 (described later) of the information processing device 100. In the case of a configuration including a display unit in the information processing device 100, the process of the display control unit 202 may be performed by the display control unit 105 of the information processing device 100.

  The information processing apparatus 100 includes a template storage unit 121, a recognition result storage unit 122, a minutes storage unit 123, a recognition unit 101, a search unit 102, a generation unit 103, a translation unit 104, a display control unit 105 And a communication control unit 106.

  The template storage unit 121 stores a summary template. 2 and 3 are diagrams illustrating an example of the summary template. As shown in FIGS. 2 and 3, the summary template includes sentences 210 and 310 (first sentence) described in Japanese (first language) and sentences 220 and 320 (320) written in English (second language). 2nd sentence). The summary template includes one or more fixed parts and a variable part. By inputting a character string that satisfies the condition into the variable section, a bilingual sentence can be generated. Note that a summary template that does not include a variable part may be used. Further, the number of languages to be associated is not limited to two, and three or more languages may be associated.

  The recognition result storage unit 122 stores the result of the voice recognition process performed by the recognition unit 101. FIG. 4 is a diagram illustrating an example of the structure of data stored in the recognition result storage unit 122. As illustrated in FIG. 4, the recognition result storage unit 122 stores information including a conference ID, a speaker, a start time of the utterance, a language, and a recognition result. The conference ID is identification information for identifying the conference. The speaker is information for identifying the user who has spoken in the conference (for example, the name of the user, the user ID, and the like). The language is information for identifying the language of the recognition result. In the example of FIG. 4, "JA" represents Japanese and "EN" represents English.

  The minutes storage unit 123 stores the minutes (minute candidates) generated by the generation unit 103. The minutes storage unit 123 stores, for example, a meeting ID and a minutes in association with each other. When a plurality of minutes is generated for a certain conference, information indicating the order of the minutes may be further stored in association with the minutes.

  In addition, each storage unit can be configured by any storage medium generally used, such as an HDD (Hard Disk Drive), an optical disk, a memory card, and a RAM (Random Access Memory). The storage unit may be a physically different storage medium, or may be implemented as physically different storage areas of the same storage medium. Further, each of the storage units may be realized by a plurality of physically different storage media.

  The recognition unit 101 receives an input of a speech signal, executes speech recognition processing on the speech signal, and converts the speech signal into text data of a recognition result. The audio signal may be input from an audio input device such as a microphone, for example, or may be input from another storage device that stores an audio signal collected in advance.

  The recognition unit 101 sequentially stores the text data as the recognition result and the information at the time of utterance in the recognition result storage unit 122 in pairs. The information at the time of the utterance includes, for example, the language at the time of the utterance, the start time of the utterance, the speaker, and the conference ID. The information at the time of the utterance may be at least information that can identify the conference and the language of the recognized text data and are arranged in time series.

  The search unit 102 searches for a summary template that matches the text data stored in the recognition result storage unit 122. The search unit 102 may start the process at an arbitrary timing, such as when a minutes generation request is received, or when the recognition unit 101 stores text data in the recognition result storage unit 122.

  For example, the search unit 102 searches the text data of the recognition result for text data that matches the summary template, and outputs the template in which the matching text data has been searched. For example, the search unit 102 extracts the text data of the recognition result for each meeting from the recognition result storage unit 122 and arranges the text data in chronological order. Then, the search unit 102 searches for a summary template that matches the information arranged in chronological order.

  The search unit 102 searches for a section that has a certain degree of similarity or more and has a character string that can be input for all variable parts in order to search for a suitable summary template. The search unit 102 searches for a section as follows, for example. First, the search unit 102 performs a morphological analysis on the text data and gives a part of speech and an original form. The search unit 102 imposes a condition that a word matching rate is equal to or higher than a certain degree for similarity, imposes a part of speech condition on a variable part, and a character string that meets the condition exists for all variable parts. Search for a section to perform. The search unit 102 determines a minimum section (a section having a minimum size) that satisfies both conditions as a section that matches the summary template.

  The processing by the search unit 102 will be described with reference to the example of FIG. In the summary template of FIG. 2, [N] and [PN] are variable parts, which means that a noun (noun phrase) and a proper noun (proper noun phrase) can be input to the variable, respectively. When this template is searched using the text data exemplified in FIG. 4, the recognition result 401 on the second line (“I do not think that the plan needs any particular correction”) and the recognition result 402 on the third line The text data corresponding to two lines (“Please proceed with the approval of the director X”) conforms to the template sentence 210 in FIG. For example, in the sentence 210, the words 211 (“[N] / wa”), the words 212 (“[PN] / no / approval /”), and the words 213 (“get”) are included in this text. Fits the data. The phrase 411 (“the plan”) can be input to the variable section [N], and the phrase 412 (“X section length”) can be input to the variable section [PN]. Therefore, the search unit 102 detects the text data corresponding to the two lines as a suitable section.

  Here, the word 413 (“correction”) is also conceivable as a possible variable, but the word 411 (“the plan / ha”) matches longer. For this reason, the search unit 102 gives priority to the phrase 411 and holds the phrase 413 (“correction”) as another candidate.

  In the above example, the section is searched for Japanese examples in the summary template, but it is also possible to search for both Japanese and English examples. Further, at the time of the search, the recognition result can be translated by the translation unit 104 described later, and the translated result can be searched in addition to the target.

  For example, when a search is performed using the text data exemplified in FIG. 4, the text data corresponding to the two lines of the recognition result 403 on the fourth line and the recognition result 404 on the fifth line match the summary template in FIG. ing. For example, the phrase 311 (“[PN] / department / no”), the phrase 312 (“sales”), and the phrase 313 (“last month”) match this text data. Also, the word 421 (“international sales”) can be input to the variable part [PN], and the word 422 (“30 percent”) and the word 423 (“11.4 million dollars”) can be input to the variable part [NUM]. . Therefore, the search unit 102 detects the text data corresponding to the two lines as a suitable section.

  At this time, for example, when the matching rate is not sufficient (for example, when the matching rate is smaller than a predetermined threshold), the translating unit 104 converts each recognition result into the other language of the summary template (when the recognition result is in Japanese). If it is, translate it into English, if English, translate it into Japanese, etc.). FIG. 5 is a diagram showing an example of the translation result translated in this manner.

  The search unit 102 can search for a matching part including such a translation result. For example, in the example of FIG. 5, the search unit 102 can determine that the word 501 (“last month”), the word 502 (“sales”), and the word 503 (“11.4 million dollars”) further match. As a result, a sufficient matching rate is obtained, and the search unit 102 detects text data corresponding to the fourth and fifth lines in FIG. 4 as a suitable section.

  It should be noted that three phrases, a phrase 422 (“30 percent”), a phrase 423 (“11.4 million dollars”), and a phrase 503 (“11.4 million dollars”) are detected as the phrases corresponding to the variable [NUM]. The search unit 102 selects, for example, the first phrase 422 found, and holds other phrases as other candidates.

  The translation unit 104 translates text data into text data in different languages. The translation unit 104 can apply any translation method such as a transfer method, an example-based method, a statistical-based method, and an intermediate language method.

  The generation unit 103 generates a minutes based on the template. For example, when a summary template that does not include the variable part is searched, the generation unit 103 generates the searched summary template as minutes. When the summary template including the variable part is searched, the generation unit 103 inputs any of the words (character strings) that can be input to the variable part to the variable part to generate the minutes.

  It should be noted that there may be a plurality of phrase candidates that can be input to the variable part, so that the character string generated first can be interpreted as a candidate for the minutes to be finally adopted. As will be described later, a candidate selected by the user from among minutes candidates is adopted as the final minutes.

  A specific example of generation of minutes (minute candidates) by the generation unit 103 will be described. For example, it is assumed that a phrase 411 that can be input to the variable part [N] and a word 412 that can be input to the variable part [PN] have been obtained for the summary template of FIG. First, the translation unit 104 translates each phrase into a corresponding language. For example, the translation unit 104 translates the Japanese words 411 and 412 into English words (“the plan” and “X director”).

  The generation unit 103 inputs the words 411, 412, and the words resulting from the translation to the corresponding variable units in the summary template. As a result, for example, a bilingual sentence including a Japanese sentence 601 in FIG. 6 (described later) and an English sentence “The plan is subject to X director's approval.” Is generated as minutes (minutes candidate).

  The same applies to the case where the variable part exists in both Japanese and English. For example, it is assumed that a phrase 421 that can be input to the variable part [PN] and a word 422 that can be input to the variable part [NUM] have been obtained for the summary template of FIG. The translation unit 104 translates the Japanese phrase 421 into an English phrase (“international marketing”), and translates the English phrase 422 into a Japanese phrase (“30%”).

  The generation unit 103 inputs each obtained phrase to a corresponding variable part in the summary template. As a result, for example, a bilingual sentence including the Japanese sentence 602 in FIG. 6 and the English sentence “Sales of international marketing division in last month were 30%.” Is generated as the minutes (minutes candidate). If another candidate exists in the variable part, the translation unit 104 performs translation for another candidate in the same manner.

  The display control unit 105 controls display of information for generating minutes. In the present embodiment, the display control unit 105 controls the display of information on the display unit 221 of the terminal device 200. The display control unit 105 transmits an instruction regarding display on the display unit 221 to the display control unit 202 via the communication control unit 106 and the communication control unit 201. In the case of a configuration including a display unit in the information processing device 100, the display control unit 105 may directly transmit an instruction regarding display to the display unit.

  The communication control unit 106 controls communication with an external device such as the terminal device 200. For example, the communication control unit 106 transmits information for displaying the generation screen to the terminal device 200. Further, the communication control unit 106 receives, from the terminal device 200, information specified by the user on the generation screen. The information specified by the user is, for example, information indicating a phrase selected as a value to be input to the variable portion, and information indicating a minutes candidate whose registration has been instructed as a final minutes among minutes candidates. And so on.

  Each of the above units (the recognition unit 101, the search unit 102, the generation unit 103, the translation unit 104, the display control unit 105, the communication control unit 106, and the like) is realized by, for example, one or a plurality of processors. For example, each of the above units may be realized by causing a processor such as a CPU (Central Processing Unit) to execute a program, that is, by software. Each of the above units may be realized by a processor such as a dedicated IC (Integrated Circuit), that is, hardware. Each of the above units may be realized by using software and hardware together. When a plurality of processors are used, each processor may realize one of the units, or may realize two or more of the units.

  Next, details of the display processing by the display control unit 105 will be described. The display control unit 105 displays, for example, a generation screen for generating minutes. FIG. 6 is a diagram illustrating an example of the generation screen. As shown in FIG. 6, the display control unit 105 displays a generation screen 600 in which a meeting log and minutes candidates are associated with each other. The conference log represents speech in the conference. For example, information in which text data, which is the recognition result of an utterance, is arranged in time series is displayed as a conference log. As shown in FIG. 6, the speaker and the text data may be displayed in association with each other.

  If it is known in advance that the language used by the user who creates the minutes is Japanese, only the Japanese sentence among the bilingual sentences generated as minutes candidates may be displayed. A bilingual document including sentences in multiple languages may be displayed.

  When a plurality of candidates exist in the variable portion of the bilingual sentence, the display control unit 105 displays each candidate in a selectable manner, and accepts the selection of the candidate. FIG. 7 is a diagram illustrating an example of a generation screen that displays a plurality of candidates in a selectable manner. For example, if the user selects another candidate 701 ("$ 11.4 million"), the variables of the minutes candidate are changed, and the corresponding translated text is also changed at the same time. FIG. 8 is a diagram illustrating an example of the generation screen after changing the value of the variable. As shown in FIG. 8, a generation screen including the changed candidate 801 is displayed.

  The generation screen further includes a function for editing minutes candidates, a function for specifying candidates to be adopted among minutes candidates, a function for specifying that the specified candidates are registered as final minutes, and the like. You may.

  For example, the generation unit 103 may generate a minutes candidate whose registration has been instructed on the generation screen as a final minutes, and store the minutes in the minutes storage unit 123. For example, the generation unit 103 receives a selection of a translated text by the user via the display control unit 105. The generation unit 103 stores the received bilingual sentences in the minutes storage unit 123 in order. The generation unit 103 generates minutes from the stored bilingual sentence list. FIG. 9 is a diagram illustrating an example of the generated minutes.

  Next, each process performed by the information processing apparatus 100 according to the first embodiment configured as described above will be described. The whole process is divided into three partial processes: a speech recognition process, a minutes candidate generation process, and a minutes generation process. Hereinafter, each process will be described.

  FIG. 10 is a flowchart illustrating an example of the voice recognition process according to the first embodiment. The voice recognition processing is started, for example, when a conference is started and when a user instructs.

  The recognizing unit 101 determines whether a voice (voice signal) has been input (step S101). When the input has been made (step S101: Yes), the recognition unit 101 performs a recognition process on the voice (step S102). The recognition unit 101 stores the recognition result and the information at the time of speech in the recognition result storage unit 122 in association with each other (step S103). The recognition unit 101 returns to step S101 and waits for the next voice input.

  When no voice is input (Step S101: No), the recognition unit 101 determines whether or not the user has instructed to end the voice recognition process (Step S104). When the end is not instructed (Step S104: No), the recognition unit 101 returns to Step S101 and repeats the processing. When the end is instructed (Step S104: Yes), the recognition unit 101 ends the voice recognition processing.

  FIG. 11 is a flowchart illustrating an example of minutes candidate generation processing according to the first embodiment. The minutes candidate generation process may be performed, for example, in parallel with the progress of the conference, that is, in synchronization with the conference, or may be performed asynchronously. For example, the minutes candidate generation process may be executed immediately after the end of the meeting or when a minutes generation request is received from the user. The minutes generation request may include the conference ID to determine which minutes of the conference to generate.

  Using the conference ID of the requested conference, the search unit 102 reads a list of recognition results of the corresponding conference and information at the time of speech (hereinafter, referred to as a list T) from the recognition result storage unit 122 (step S201). . The search unit 102 generates information in which the list T is arranged in chronological order (hereinafter, referred to as information T ') (step S202).

  The search unit 102 acquires one unsearched summary template from the template storage unit 121 (step S203). Further, the search unit 102 prepares a detection list Y for storing search results.

  The search unit 102 searches the information T 'for a section that matches the acquired summary template (step S204). If there is a matching section (hereinafter, referred to as section R) (step S204: Yes), the search unit 102 determines whether there is a section R, a matching template (hereinafter, referred to as template U), and a variable section. A set of a character string that can be input to the variable part is added to the detection list Y (step S205).

  After the addition to the detection list Y and when there is no suitable section (step S204: No), the search unit 102 determines whether or not all the summary templates have been processed (step S206). If all the summary templates have not been processed (step S206: No), the search unit 102 returns to step S203 and repeats the process for the next unprocessed summary template.

  When all the summary templates have been processed (Step S206: Yes), the generation unit 103 generates minutes candidates from the summary templates included in the obtained detection list Y (Step S207). For example, when there is a variable part, the generation unit 103 inputs any of the candidates that can be input to the variable part to the variable part, and generates a bilingual sentence. The generating unit 103 generates a bilingual sentence list P, which is a list of bilingual sentences corresponding to each of the summary templates, as minutes candidates.

  Next, the display control unit 105 displays the information T 'and the bilingual sentence list P in association with each other (step S208). For example, the display control unit 105 displays information T ′ (meeting log) and a bilingual sentence list P (minute candidates) using a generation screen 600 as shown in FIG.

  FIG. 12 is a flowchart illustrating an example of the minutes generation process according to the first embodiment. The minutes generation process is performed after the minutes candidate generation process. For example, the minutes generation process is appropriately executed immediately after the minutes candidate generation process, or when explicitly requested by the user. In the following, an example will be described in which the minutes generation processing is executed after the generation screen 600 shown in FIG. 6 is displayed, for example.

  The generation unit 103 determines whether the user has instructed to change the variable unit (Step S301). When the change is instructed (Step S301: Yes), the generation unit 103 changes the variable part to the instructed value (phrase) (Step S302). The display control unit 105 updates and displays the generation screen so as to include the translated text in which the change is reflected.

  After changing the value, or when the change is not instructed (step S301: No), the generation unit 103 determines whether or not the addition of the minutes is instructed (step S303). When addition is instructed (step S303: Yes), the generation unit 103 generates a bilingual sentence corresponding to the added minutes and adds the bilingual sentence to the minutes of the corresponding meeting in the minutes storage unit 123 (step S303). S304).

  After adding the minutes, or when the addition is not instructed (step S303: No), the generation unit 103 determines whether or not the end of the process is instructed (step S305). When the end is not instructed (step S305: No), the generation unit 103 returns to step S301 and repeats the processing. When the end is instructed (Step S305: Yes), the generation unit 103 ends the minutes generation process.

  As described above, the information processing apparatus according to the first embodiment searches the template of the summary (minutes) for a template that matches the speech recognition result, and generates the minutes using the searched template. As a result, it is possible to easily create necessary and sufficient contents for generating the summary. In addition, since the template is associated with sentences in a plurality of languages, it is possible to simultaneously provide summaries in multiple languages.

(Second embodiment)
In the information processing apparatus according to the second embodiment, minutes editing processing is added to the processing of the first embodiment. Since the speech recognition process, the minutes candidate generation process, and the minutes generation process are the same, the description is omitted.

  FIG. 13 is a block diagram illustrating a configuration example of an information processing device 100-2 according to the second embodiment. As shown in FIG. 13, the information processing apparatus 100-2 includes a template storage unit 121, a recognition result storage unit 122, a minutes storage unit 123, a recognition unit 101, a search unit 102, a generation unit 103, It includes a translation unit 104, a display control unit 105, a communication control unit 106, and an editing unit 107-2.

  The second embodiment is different from the first embodiment in that an editing unit 107-2 is added to the information processing apparatus 100-2. Other configurations and functions are the same as those in FIG. 1 which is a block diagram of the information processing apparatus 100 according to the first embodiment, and therefore, are denoted by the same reference numerals and description thereof will be omitted.

  The editing unit 107-2 edits the generated summary (minutes). For example, the editing unit 107-2 acquires the minutes from the minutes storage unit 123, and edits the minutes according to the editing operation by the user. The editing unit 107-2 changes the order of the translated sentences, corrects the variable part, adds an arbitrary sentence, and deletes the minutes.

  FIG. 14 is a diagram illustrating an example of the editing operation. FIG. 14 shows a part of the sentence of the first line in Japanese (“the plan must be approved by the director of X”) in the bilingual sentence of FIG. 9 as a phrase 1401 (“●● project plan”). An example in which a new Japanese sentence 1402 (“a 10% increase in sales is expected this month.”) Is added.

  These editing operations are performed in the language used by the minutes creator (in this case, Japanese). After that, the editing unit 107-2 translates the changed portion using the translation unit 104, and inputs the translated portion into the corresponding portion of the minutes. As a result, the phrase 1411 (“The plan for ●● project”) in the English sentence is also changed, and the English sentence 1412 (“This month we further expected to increase in sales of 10% ") Is added.

  Next, a minutes editing process performed by the information processing apparatus 100-2 according to the second embodiment thus configured will be described with reference to FIG. FIG. 15 is a flowchart illustrating an example of the minutes editing process according to the second embodiment. The minutes editing process is started, for example, after an explicit request from a user or the end of the minutes generating process.

  The editing unit 107-2 receives the meeting ID of the meeting whose minutes are to be edited, acquires the minutes of the corresponding meeting, and displays it by the display control unit 105 (step S401). Thereafter, the editing unit 107-2 waits for an instruction from the user.

  The editing unit 107-2 determines whether editing has been instructed (step S402). When the editing is instructed (Step S402: Yes), the editing unit 107-2 edits the minutes according to the instruction (Step S403).

  For example, when the editing of the variable portion of the bilingual sentence is instructed, the editing unit 107-2 updates the variable portion according to the instruction. When an instruction to add an arbitrary sentence is instructed, the editing unit 107-2 passes the sentence to the translating unit 104, causes the added sentence to be translated, and adds the translated sentence to the added sentence. To be added to the minutes. When an instruction to change the order of the text or to delete the text is issued, the editing unit 107-2 executes the text replacement or deletion according to the instruction.

  After the editing is completed, or when the editing is not instructed (step S402: No), the editing unit 107-2 determines whether the end of the process is instructed (step S404). If termination has not been instructed (step S404: No), the editing unit 107-2 returns to step S402 and repeats the processing.

  When the end is instructed (step S404: Yes), the editing unit 107-2 registers (stores) the minutes edited so far in the minutes storage unit 123 (step S405), and executes the minutes editing process. finish.

  As described above, in the information processing apparatus according to the second embodiment, the generated minutes can be further edited.

  As described above, according to the first and second embodiments, it is possible to create minutes (summary) in multiple languages more smoothly. The information processing apparatus, the information processing method, and the program according to the first and second embodiments are suitable for, for example, translating spoken words and transmitting meaning.

  Next, a hardware configuration of the information processing apparatus according to the first or second embodiment will be described with reference to FIG. FIG. 16 is an explanatory diagram illustrating an example of a hardware configuration of the information processing apparatus according to the first or second embodiment.

  An information processing apparatus according to the first or second embodiment includes a control device such as a CPU (Central Processing Unit) 51, a storage device such as a ROM (Read Only Memory) 52 and a RAM (Random Access Memory) 53, a network, And a communication I / F 54 for performing communication by connecting to each other, and a bus 61 for connecting each unit.

  A program executed by the information processing apparatus according to the first or second embodiment is provided by being incorporated in the ROM 52 or the like in advance.

  The program executed by the information processing apparatus according to the first or second embodiment is a file in an installable format or an executable format, which is a CD-ROM (Compact Disk Read Only Memory), a flexible disk (FD), or a CD. -It may be configured such that it is recorded on a computer-readable recording medium such as R (Compact Disk Recordable) or DVD (Digital Versatile Disk) and provided as a computer program product.

  Further, a program executed by the information processing apparatus according to the first or second embodiment is stored on a computer connected to a network such as the Internet, and provided by being downloaded via the network. Is also good. Further, the program executed by the information processing apparatus according to the first or second embodiment may be provided or distributed via a network such as the Internet.

  The program executed by the information processing apparatus according to the first or second embodiment can cause a computer to function as each unit of the information processing apparatus described above. In this computer, the CPU 51 can read out a program from a computer-readable storage medium onto a main storage device and execute the program.

  While some embodiments of the invention have been described, these embodiments have been presented by way of example only, and are not intended to limit the scope of the inventions. These new embodiments can be implemented in other various forms, and various omissions, replacements, and changes can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope and gist of the invention, and are also included in the invention described in the claims and their equivalents.

REFERENCE SIGNS LIST 100, 100-2 information processing device 101 recognition unit 102 search unit 103 generation unit 104 translation unit 105 display control unit 106 communication control unit 107-2 editing unit 121 template storage unit 122 recognition result storage unit 123 minutes storage unit 200 terminal device 201 communication control unit 202 display control unit 221 display unit

Claims (6)

A first sentence described in a first language and a second sentence described in a second language are associated with one or more text data described in one or more languages obtained by recognizing a voice. A search unit that searches for text data that matches the template of the abstract, and outputs the template where the matching text data is searched;
A display control unit that displays the output template in association with the text data that matches,
A generation unit that generates a summary based on the template ,
The first sentence and the second sentence include a variable into which a value can be input,
The generating unit generates a summary by inputting a phrase included in the text data that is adapted to a variable included in the first sentence and the second sentence,
The display control unit, when a plurality of words that can be input to the variable is included in the text data, display a plurality of words in a selectable manner,
The generating unit generates a summary by inputting a selected phrase among the plurality of displayed phrases to the variable,
Information processing apparatus. The apparatus further comprises a translation unit that translates the text data into text data in different languages,
The search unit searches the text data to which the translated text data has been added, for text data that matches the summary template.
The information processing device according to claim 1. Further comprising an editing unit for editing the generated summary,
The information processing device according to claim 1. It further includes a recognition unit that recognizes voice and outputs text data,
The search unit searches for text data matching the template from among the one or more text data output by the recognition unit, and outputs the template in which the matching text data has been searched.
The information processing device according to claim 1. A first sentence described in a first language and a second sentence described in a second language are associated with one or more text data described in one or more languages obtained by recognizing a voice. A search step of searching for text data that matches the template of the abstract, and outputting the template in which the matching text data has been searched;
A display control step of displaying the output template in association with the matching text data,
Generating a summary based on the template ,
The first sentence and the second sentence include a variable into which a value can be input,
The generating step generates a summary by inputting a phrase included in the text data that is adapted to a variable included in the first sentence and the second sentence,
The display control step, when a plurality of words that can be input to the variable is included in the text data, display a plurality of words in a selectable manner,
The generating step generates a summary by inputting a selected phrase among the plurality of displayed phrases to the variable,
Information processing method. Computer
A first sentence described in a first language and a second sentence described in a second language are associated with one or more text data described in one or more languages obtained by recognizing a voice. A search unit that searches for text data that matches the template of the abstract, and outputs the template in which the matching text data is searched;
A display control unit that displays the output template in association with the text data that matches,
A generating unit that generates a summary based on the template ,
The first sentence and the second sentence include a variable into which a value can be input,
The generating unit generates a summary by inputting a phrase included in the text data that is adapted to a variable included in the first sentence and the second sentence,
The display control unit, when a plurality of words that can be input to the variable is included in the text data, display a plurality of words in a selectable manner,
The generating unit generates a summary by inputting a selected phrase among the plurality of displayed phrases to the variable,
Program.

Images