JP4589910B2 - Conversation recording blogging device - Google Patents

Description

  The present invention relates to a conversation recording blogging apparatus. More particularly, the present invention relates to a support apparatus and method for creating a blog using a mobile phone, a portable recording device, or the like.

  Recently, a simple homepage called a so-called blog (abbreviation of webblog) is rapidly spreading on the Internet. Blogs are not just personal diary homepages, but are often used to express analysis and opinions rooted in your expertise and position, and discuss with other site authors about current news and specialized topics. The role of various communication tools, knowledge management tools, and advertising media is increasing.

  The features of the blog are as follows: (1) Each article has a time record (timestamp), (2) Since articles are arranged in time series, the latest article is always located at the top, (3) Each article can be linked ( 4) Mutual communication is possible through links and comments. Based on these characteristics, the number of creators and viewers of blogs is increasing rapidly.

In order to create a blog, a method of inputting a sentence to be a blog article and uploading it together with an image such as a photo on a PC or a mobile phone is generally used. However, every day, every day, every time an event occurs, it's quite annoying for blog creators to enter and upload text. For this reason, tools for creating and updating daily blogs as easily as possible have appeared. For example, in Patent Literature 1, when a user performs a specific action, the position information of the own device (mobile communication device) is acquired, and the event information of the user is stored from the date and time of the action and the position information. A portable communication device is disclosed in which a user easily creates a blog by transmitting event information selected by the user later to a server.
JP 2006-301844 A

  However, in a method such as Patent Document 1, a user designates a specific action (for example, when using electronic money, an electronic commuter pass, an electronic ticket, etc., or when captured by a camera) as an event in advance. It must be set and no events other than the specified specific action will be recorded. In addition, it is necessary to store and accumulate a large number of these events, and it is troublesome to specify ON / OFF whether to automatically generate a comment to be uploaded to the blog for each item of these events.

  In view of the above-described problems, the present invention uses a mobile terminal device having a conversation recording unit such as a mobile phone or a mobile recorder (IC recorder or the like), and more easily obtains information for making a blog. An object of the present invention is to provide a new apparatus and method.

  The present invention provides the following solutions.

(1) A device for creating the main points of conversation recording using voice conversation recording,
A conversation record input unit that accepts a conversation record stored in the user's mobile terminal as an input;
A conversation separating unit that separates the voice data of the conversation record into an utterance part of the user and an utterance part of another person talking with the user;
A voice text conversion unit for converting the separated voice data into text data;
A key word extraction unit that extracts key words representing features of conversation from the text data;
A display unit for displaying the extracted key words to the user;
A device comprising:

  According to such a configuration, this apparatus stores conversation records (call recording and conversation recording) stored in a user's portable terminal (for example, a portable terminal having a voice recording unit such as a cellular phone or an IC recorder). Accepted as input, only the voice data portion (speech portion) of the conversation record is separated and extracted. Speech text conversion is performed on the separated speech data, and “main point words” (main point keywords) representing the characteristics of the recorded conversation are extracted from the converted text data. Even if the speech-to-speech conversion is incomplete, it is possible to extract only the key words that appear with high frequency (actually weighted frequency) in the conversation. This makes it possible to extract the key words included in the user's daily recording based on the conversation recording stored in the portable terminal by voice and display it as text in time series to the user. Such key word text can be used for various purposes later.

(2) A device for supporting the creation of a blog using voice conversation recording,
A conversation record input unit that accepts a conversation record stored in the user's mobile terminal as an input;
A conversation separating unit that separates the voice data of the conversation record into an utterance part of the user and an utterance part of another person talking with the user;
A voice text conversion unit for converting the separated voice data into text data;
A key word extraction unit that extracts key words representing features of conversation from the text data;
A display unit for displaying the extracted key words to the user;
A blog data editing unit that allows the user to edit a comment to be posted on a blog based on the gist word;
A blog data transmission unit that transmits blog data including the comment to a server;
A device comprising:

  According to such a configuration, this apparatus stores conversation records (call recording and conversation recording) stored in a user's portable terminal (for example, a portable terminal having a voice recording unit such as a cellular phone or an IC recorder). Accepted as input, only the voice data portion (speech portion) of the conversation record is separated and extracted. Speech text conversion is performed on the separated speech data, and “main point words” (main point keywords) representing the characteristics of the recorded conversation are extracted from the converted text data. The extracted key words are presented to the user and used as original data for creating a comment for posting on the blog. Then, a means for allowing the user to edit the original data is uploaded to the web server of the blog site as a blog article (blog comment).

  Thus, the daily record of the user can be displayed to the user by extracting the key words included in the record based on the conversation record stored in the portable terminal. In addition, in some cases, while listening to the conversation record, you can recall the events of the day and reflect them in the blog post. Since these conversation records are stored in time series, timestamp information for each article, which is a feature of the blog, can be easily extracted. In addition to the personal thoughts and impressions of the user, information from the conversation partner (call partner or interview record partner) can also be extracted at the same time.

  (3) The apparatus according to (1) to (2), wherein the conversation record is a call record using a mobile phone.

  According to such a configuration, if a mobile phone widely used today as a conversation record is provided with a call record, it can be used as an information source for the conversation record, and the above function can be realized.

  (4) The apparatus according to (1) to (2), wherein the conversation recording is conversation recording data recorded in the portable terminal.

  According to such a configuration, the functions (1) to (2) are based on a record intentionally recorded by a user, for example, a record of an interview or interview for a specific event, instead of a call record of a mobile phone. Of course, if the mobile phone has a recording function in addition to call recording, call recording and recording can be realized with a single mobile phone.

  (5) The apparatus according to any one of (1) to (4), wherein the extraction of the key word is obtained by calculating a TFIDF value from the text data.

  According to such a configuration, it is possible to extract a key word representing the characteristics of the conversation using the TFIDF method from a call record or a recording record that has been converted into a voice text. TFIDF will be described later.

(6) The apparatus according to (5), wherein the TFIDF value is calculated for each of the text data of the utterance part of the user and the text data of the utterance part of the other party of the conversation, or the whole of both data. .

  According to such a configuration, the fact that both the user's own utterance content and the utterance content of the other party (calling partner or interviewing partner) are recorded in the conversation record can be used as a key word from two different aspects. Can be extracted. For example, the range of selection of information to be used, such as extracting key words using only the user's own utterance part, using only the other party's utterance record, or using both together, is widened. It will be.

  (7) The apparatus according to (2), wherein the blog data editing unit has a function of editing a display format of the blog, and transmits the blog data to the server according to an instruction from the user.

  According to such a configuration, the device has a function of editing not only the blog content but also the blog display format (for example, HTML format or XML format) itself, so that the content uploaded on the device side can be previewed, It becomes possible to upload in the form posted on the blog.

  (8) The said portable terminal is an apparatus as described in (1)-(7) provided with the means to image a target object.

  According to such a configuration, for example, by providing the mobile terminal with a digital camera, it is possible to easily acquire an image to be uploaded together with the blog article.

  (9) The device according to any one of (1) to (8), wherein the portable terminal includes means for acquiring own position information, and records position information of a place where the conversation record is created in association with the conversation record. .

  According to such a configuration, the mobile terminal is provided with position information acquisition means such as GPS, and the shooting location information and the like can be easily used for the blog together with the date and time when the photo was taken.

(10) A device for supporting the creation of a blog using a voice utterance record,
An utterance record input unit that accepts an utterance record stored in the user's mobile terminal as an input;
An utterance separation unit that separates the voice data of the utterance record into the utterance part of the user and the other part;
A voice text conversion unit for converting the separated voice data into text data;
A key word extraction unit that extracts key words representing features of the utterance from the text data;
A display unit for displaying the extracted key words to the user;
A blog data editing unit that allows the user to edit a comment to be posted on a blog based on the gist word;
A blog data transmission unit that transmits blog data including the comment to a server;
A device comprising:

  Even when there is no call partner or interview partner, the function similar to (2) can be realized by using the user's own dictation record (self-recording). This can be considered as a voice memo for creating a blog.

(11) A method for creating a point of conversation recording using voice conversation recording,
In computer systems,
Receiving a conversation record stored in the user's mobile device as input,
Separating the voice data of the conversation record into an utterance part of the user and an utterance part of another person talking with the user;
Converting the separated audio data into text data;
Extracting key words representing conversation characteristics from the text data;
Displaying the extracted key words to the user;
Including methods.

  According to such a configuration, the function of the device of (1) can be regarded as a method invention in a computer system, and the same effect as that of (1) can be obtained.

  Thus, according to the present invention, it is possible to easily obtain key words for facilitating the creation of a blog using a mobile terminal device having a conversation recording unit such as a mobile phone or a mobile recording device.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a functional block diagram of a conversation recording apparatus 10 relating to blog creation according to an embodiment of the present invention. For convenience of explanation, not only the conversation recording device 10 but also a mobile phone 30, an IC recorder, and a Web server 20 connected thereto are shown at the same time. The conversation recording apparatus 10 mainly includes a conversation recording input unit 11 that accepts as input a conversation record recorded in a recording device such as a mobile phone 30 or an IC recorder 31. Conversation separation unit 12 that separates into silent parts, speech text conversion unit 13 that converts the separated conversational speech part into text, main point word extraction unit 14 that extracts a main word representing its characteristics from the converted text, A blog data editing unit 15 that allows the user to edit the blog data, and a blog data transmission unit 16 that transmits the edited blog data to the WEB server 20 that is a blog site. The conversation recording device 10 includes a general audio speaker 17, a display unit 18 (liquid crystal display device or the like) for a user, an operation unit 19 (keyboard, mouse, etc.), and a voice input if necessary. A microphone (not shown) is provided.

  Connection with the mobile phone 30 or the IC recorder 31 may be performed using a general-purpose interface such as USB, or a portable memory such as an SD card or a SMART card may be attached and detached to exchange data.

  The conversation recording apparatus 10 is connected to the Web server 20 via a network 40 (typically the Internet, but may be a local area network (LAN) or a wide area network (WAN)). The web server 20 is a server that stores the contents of a blog, a blog data receiving unit 21 that receives data from the conversation recording device 10, a blog recording unit 22 that edits data into a web page format, and a web page DB 23 that stores data. Is provided.

  Note that these configurations are merely examples, and other configurations may be included as long as they are functionally equivalent, or may be replaced with the original configurations.

  FIG. 2 is a diagram showing a processing flow of the conversation recording apparatus 10 according to one embodiment of the present invention. Here, the case of the cellular phone 30 will be described as an input, but the basic flow is the same even in the case of other recording devices such as the IC recorder 31.

  First, in step S <b> 1, a call record is acquired from the mobile phone 30. As described above, this data acquisition may be performed by connecting the conversation recording device 10 and the mobile phone 30 with a USB or a dedicated cable, or may use an SD memory card or the like.

  Next, in step S2, the utterance part of the user (blog creator) itself, the utterance part of the other party of the call, and the silent part are separated. As the separation technique itself, any known technique may be used. The reason why the entire conversation is separated into the part of the person and the other party is not only to reduce the misconversion rate of the speech recognition (next step S3) but also to the thinking of the person and the other party. This is also to make it easier to analyze differences in terminology due to differences in patterns. The reason for removing the silent portion is to reduce the time when the recorded content is reproduced later.

  Next, in step S3, the separated voice data is converted into text. For this conversion, that is, voice recognition, a known technique (for example, see Japanese Patent Laid-Open No. 23030295) may be used. However, in consideration of the presence of a plurality of speakers, a voice pattern and a dictionary can be switched among a plurality. preferable.

  Next, in step S4, the converted text is divided into morphemes. A morpheme is the smallest linguistic unit having meaning, and is a linguistic unit used when a sentence written in a natural language is divided. For example, “Today is good weather” is divided into “Today / Has / Good / Weather / Is it”.

  Further, in step S5, the morpheme-divided text treats the entire text (the entire call) as a document set, and further treats an appropriate call unit (for example, a fixed time) as a document, and is a feature that characterizes the document. Extract words (keywords). The key word is not a simple appearance frequency in the whole text, but a word that appears biased in a specific document is extracted. Specifically, for example, a TFIDF value is calculated, and a word indicating that the value is higher than a predetermined value is extracted as a key word. The TFIDF method is a method of weighting keywords based on the appearance frequency of words. TFIDF is divided into TF and IDF. TF is Term Frequency (vocabulary frequency), and IDF is Inverse Document Frequency (reciprocal of document frequency). That is, TFIDF represents the product of TF and IDF. In the following embodiments, TFIDF is mainly described, but extraction of key words is well known in text mining other than TFIDF (for example, document clustering), and the use of those known techniques is excluded. Not what you want.

  In step S6, the processing in steps S2 to S5 is repeated for all recorded call records. By doing so, the recorded call record can be converted into text. If the conversion accuracy of the voice text is low, the user may check the recognized text and perform correction if necessary before performing the process of step S4.

  Next, in step S7, the key words extracted for each call record are displayed to the user. At this time, the key word may be displayed as a candidate for the title of the blog article, or the generation time of the article may be displayed from the time stamp of the call record. Moreover, you may make it display the photograph image | photographed with the camera and the positional information taken in from GPS. A blog editing screen for a user to create a blog article based on these various types of information is displayed. It is also possible to display text-converted call records at the same time and use it as the original blog article. In this way, various kinds of information for creating a blog article can be obtained by the conversation recording apparatus 10 of the present invention.

  Finally, in step S8, the edited blog data is transmitted (uploaded) to the Web server based on the user's instruction. The uploaded blog data is processed into a format (HTML conversion or the like) that is easy to browse on the blog site and stored in the Web page DB 23.

  FIG. 3 is a diagram showing an example of a call record in the mobile phone 30 as a conversation record. The call record shown in the figure is the content of a mobile phone call that user (blog creator) A exchanged with friend B. Here, the topic of the cake shop where Mr. A stopped by accident is exciting, but it is often done to upload such a memorable event as a blog post later. At this time, since the call record includes a timestamp, it is very convenient for creating a blog to edit this in time series. Also, many of today's mobile phones are equipped with a camera and GPS (Global Positioning System), and it is useful for uploading to the blog the images taken by the camera, the time and position, etc. Information. For example, if you post an impression of a cake you happened to eat at that time, its photo, or the location information (map) of the store on a blog, it is quite possible that many people will see it. obtain.

  FIG. 4 is a diagram showing a table in which key data (collection of key words) of the call record is extracted based on the call record shown in FIG. In this table, the words (keywords) extracted from the above-mentioned call record are stored together with the appearance frequency and the value of TFIDF, and the words with higher TFIDF values are arranged at the top as the key words representing the features. Here, the TFIDF value is an example in which the entire call is calculated as a document set. However, the TFIDF value may be shown for each of the utterance portions of the user (Mr. A) and the friend (Mr. B). For example, if the terms used by Mr. A and Mr. B are different in the same meaning, the results of the key words obtained from Mr. A's speech part and the key words obtained from Mr. B's speech part are naturally different. By analyzing this, when creating a blog article, it is possible to expand the scope of selection such as taking in the opinions of the other party of the conversation or not taking them in reverse. As a result, it is possible to analyze variations in terms used by Mr. A and Mr. B. Further, instead of making the entire call into one document, an appropriate break (for example, when the silent time lasts long) may be detected and treated as a separate document for each break. Alternatively, a call may be divided at a certain time, for example, every 3 minutes, and each may be handled as one document. Thus, since various document units can be defined even from the same call record, the TFIDF value can be calculated for each and used to analyze the key words.

  FIG. 5 is a diagram showing recording recording (interview) in a recording machine such as the IC recorder 31 as conversation recording. In this example, conversations are recorded when Mr. A and Mr. B visit an actual store and interview the store manager after a call on the mobile phone 30 of FIG. Unlike a call on the mobile phone 30, the interview is considered to be conducted on the assumption that the interview is posted on the blog, so there is a high possibility that a record that is likely to become a blog article is obtained. The conversation record obtained here is input to the conversation recording device 10 in the same manner as the call record of the mobile phone 30, and is converted into voice text to be the original data of the blog article. It should be noted that such interview recording can be substituted by the mobile phone 30 as long as it has a recording function, even if it is not a dedicated recording machine such as the IC recorder 31. Even if there is no other party, the user can dictate the article, and the dictation record can be used as the input of the conversation recording apparatus 10.

  FIG. 6 is a diagram showing a table in which the key data (collection of key words) of the recording record is extracted based on the recording record of FIG. In this table, as in the table of FIG. 4 described above, words (keywords) extracted from the recording are stored together with the appearance frequency and the value of TFIDF, and the higher the TFIDF value, the higher the key words representing the features. Are lined up. The calculation method of the TFIDF value is also the same. However, if the contents of the call and the recorded contents are related as shown in this example, the call record is document 1 and the recorded record is document 2, and the TFIDF value is calculated from the entire documents, and based on that value. Key words can be extracted. Of course, the TFIDF values may be obtained separately for the document 1 and the document 2, and the key words may be extracted based on these values. In this way, it is possible to extract key words from different angles. For example, when the conversation partner does not speak useful information and is just talking, the utterance part of the other party can be excluded from the extraction target.

  FIG. 7 is a diagram showing an example of a blog article editing screen in the conversation recording apparatus 10 according to one embodiment of the present invention. Here, a case of call recording of a mobile phone will be described. A target call record or the like is selected from a call record selection bar indicated by reference numeral 50. A plurality of call records may be selected. The area 51 displays a call start time, a call partner, a call time, and the like as information related to the selected call. Further, in the area 52, the key words (key points keywords) extracted based on the above-described method are displayed. Here, the TFIDF values may be displayed in descending order. In the area 53, the voice recognition result of the call record converted into text by the voice text conversion unit 13 is displayed. In this text, the underlined portion indicates a misrecognized or unrecognized portion. The user does not need to particularly correct the words that are not correctly recognized. In the area 54 (blog article heading section), for example, the call-related information in the area 51 and the title part of the blog article created from the key point keywords in the area 52 are displayed. Of course, the user may edit the title portion. The area 55 is an area for displaying the extracted key word as a key point of the content (comment) of the blog article. Of course, this portion may be edited by the user based on these key words, or in some cases, the call record in the area 53 can be directly edited.

  Regions 56 to 59 are for editing images such as photos that are simultaneously uploaded to the blog. That is, a photo file is selected from the photo selection bar in the area 56, and the shooting time and location are displayed in the area 57 (photo attached information). When a mobile phone equipped with GPS is used, location information on a place may be acquired. In the area 58, various tool buttons and the like for editing a photograph are arranged so that the photograph to be uploaded is displayed in the area 59. Although not shown, the photo before editing and the photo after editing may be displayed side by side.

  The images of the blog article created in this way and images such as photos are selected by the user by selecting an upload destination blog in the area 60 (blog selection menu) and pressing the upload button 62. It is published in. Further, by providing a function for editing a blog display format (for example, HTML or XML), a browsing image may be displayed by pressing the preview button 61 before uploading.

[Hardware configuration of conversation recording device]
FIG. 8 is a diagram illustrating an example of a hardware configuration of the conversation recording apparatus 10 according to an embodiment of the present invention.

  The conversation recording apparatus 10 includes a central processing unit (CPU) 110 (a plurality of CPUs such as the CPU 120 may be added in a multiprocessor configuration) 110, a bus line 105, a communication I / F 140, a main memory 150, A BIOS (Basic Input Output System) 160, a USB port 190, an I / O controller 170, and input means such as a keyboard and mouse 180 and a display device 122 are provided. Further, a microphone / speaker 173 may be provided as necessary. The mobile phone 30 and the IC recorder 31 may be connected to the USB port 190 so that the call recording data can be taken into the conversation recording device 10.

  Storage means such as a tape drive 172, a hard disk 174, an optical disk drive 176, and a semiconductor memory 178 can be connected to the I / O controller 170.

  The BIOS 160 stores a boot program executed by the CPU 110 when the conversation recording apparatus 10 is activated, a program depending on the hardware of the conversation recording apparatus 10, and the like.

  The hard disk 174 constituting the storage unit 102 stores various programs for the conversation recording device 10 to function as a device and programs for executing the functions of the present invention, and can further configure various databases as necessary. .

  As the optical disk drive 176, for example, a DVD-ROM drive, a CD-ROM drive, a DVD-RAM drive, or a CD-RAM drive can be used. In this case, the optical disk 177 corresponding to each drive is used. A program or data can be read from the optical disk 177 by the optical disk drive 176 and provided to the main memory 150 or the hard disk 174 via the I / O controller 1070. Similarly, the tape medium 171 corresponding to the tape drive 172 can be used mainly for backup.

  The program provided to the conversation recording device 10 is provided by being stored in a recording medium such as the hard disk 174, the optical disk 177, or a memory card. The program may be installed in the conversation recording apparatus 10 and executed by being read from the recording medium via the I / O controller 170 or downloaded via the communication I / F 140.

  The aforementioned program may be stored in an internal or external storage medium. Here, in addition to the hard disk 174, the optical disk 177, or the memory card, a magneto-optical recording medium such as an MD, or a tape medium can be used as the storage medium constituting the storage unit 102. Further, a storage device such as a hard disk 174 or an optical disk library provided in a server system connected to a dedicated communication line or the Internet may be used as a recording medium, and the program may be provided to the conversation recording device 10 via the communication line.

  Here, the display device 122 displays a screen for accepting data input to the user or displays a screen of a calculation processing result by the conversation recording device 10, and is a cathode ray tube display device (CRT) or a liquid crystal display device. (LCD) and other display devices.

  Here, the input means accepts input by the user, and may be configured by a keyboard, a mouse 180, and the like.

  The communication I / F 140 is a network adapter for enabling the conversation recording apparatus 10 to be connected to a terminal via a dedicated network or a public network. The communication I / F 140 may include a modem, a cable modem, and an Ethernet (registered trademark) adapter.

  In the above example, the conversation recording apparatus 10 has been mainly described. However, the functions described above can be realized by installing a program in a computer and operating the computer as a server apparatus. Therefore, the functions realized by the server described as an embodiment in the present invention can be realized by executing the above-described method by the computer, or by introducing the above-mentioned program into the computer and executing it. It is.

  As mentioned above, although embodiment of this invention was described, this invention is not restricted to embodiment mentioned above. The effects described in the embodiments of the present invention are only the most preferable effects resulting from the present invention, and the effects of the present invention are limited to those described in the embodiments of the present invention. is not.

It is the figure which showed the functional block of the conversation recording apparatus 10 which concerns on one Embodiment of this invention. It is the figure which showed the processing flow of the conversation recording apparatus 10 which concerns on one Embodiment of this invention. It is the figure which showed an example of the call recording in a mobile telephone as conversation recording. It is the figure which showed the table which extracted the key data (collection of key words) of the call record based on the call record of FIG. It is the figure which showed the recording recording (interview) in recording machines like an IC recorder as conversation recording. It is the figure which showed the table which extracted the key data (collection of key words) of the sound recording based on the sound recording of FIG. It is the figure which showed an example of the blog article edit screen in the conversation recording apparatus 10 which concerns on one Embodiment of this invention. It is a figure which shows an example of the hardware constitutions of the conversation recording apparatus 10 which concerns on one Embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Conversation recording apparatus 11 Conversation record input part 12 Conversation part isolation | separation part 13 Speech text conversion part 14 Gist word extraction part 15 Blog data edit part 16 Blog data transmission part 17 Speaker 18 Display part 19 Operation part 20 Web server 21 Blog data reception part 22 Blog recording section 23 Web page DB
30 Cellular Phone 31 IC Recorder 40 Network 50 Call Record Selection Bar 51 Call Related Information 52 Extraction Key Words 53 Call Record (After Text Conversion)
54 Blog Article Heading 55 Blog Article Content 56 Photo Selection Bar 57 Photo Accompanying Information 58 Photo Editing Tool Buttons 59 Photos 60 Blog Selection Menu 61 Review Button 62 Upload Button

Claims (8)

A device for supporting the creation of a blog using voice conversation recording,
A voice text conversion unit that converts voice conversation recording data stored in the portable terminal into text data by voice recognition ;
The text data is divided into morphemes, the entire conversation in a predetermined range is made into a document set, a predetermined conversation unit is taken as a document, a TFIDF value is calculated, and the characteristic of the conversation is determined for a word having a higher calculated TFIDF value. A keyword extractor for extracting as a keyword to represent,
The keywords that the extracted, the user of the terminal, and a display unit that displays along with the blog data editing screen to edit the comments posted on the blog,
A blog data transmission unit that transmits blog data including the comment to a server;
A device comprising: The apparatus according to claim 1 , wherein the voice conversation record is a call record using a mobile phone. The voice data of the voice conversation record is further provided with a conversation separating unit that separates the utterance part of the user and the utterance part of another person talking with the user by a difference in pronunciation,
The keyword extraction unit calculates the TFIDF value by regarding the entire conversation within a predetermined range of the text data of the user's utterance part and the text data of the utterance part of the conversation partner as a document set. The apparatus according to claim 2 . The apparatus according to claim 1 , wherein the blog data editing screen has a function of editing a display format of the blog, and transmits the blog data editing screen to the server according to an instruction from the user. The mobile terminal comprises means for capturing an object, according to claims 1 to 4. The apparatus according to any one of claims 1 to 5 , wherein the portable terminal includes means for acquiring own position information, and records position information of a place where the voice conversation record is created in association with the voice conversation record. A device for supporting the creation of a blog using voice utterance records,
A voice text conversion unit that converts voice utterance recording data stored in a portable terminal into text data by voice recognition ;
The text data is divided into morphemes, a whole set of utterances in a predetermined range is taken as a document set, a predetermined utterance unit is taken as a document, a TFIDF value is calculated, and the higher the calculated TFIDF value, the characteristics of the utterance A keyword extractor for extracting as a keyword to represent,
The keywords that the extracted, the user of the terminal, and a display unit that displays along with the blog data editing screen to edit the comments posted on the blog,
A blog data transmission unit that transmits blog data including the comment to a server;
A device comprising: A method for supporting the creation of a blog using voice conversation recording,
In computer systems,
Converting voice conversation recording data stored in the mobile terminal into text data by voice recognition ;
The text data is divided into morphemes, the entire conversation in a predetermined range is made into a document set, a predetermined conversation unit is taken as a document, a TFIDF value is calculated, and the characteristic of the conversation is determined for a word having a higher calculated TFIDF value. Extracting as keywords to represent,
The keywords that the extracted, the user of the terminal, and a display step of displaying along with the blog data editing screen to edit the comments posted on the blog,
A blog data transmission step of transmitting blog data including the comment to a server;
Including methods.

Images