JP2013172381A - Content storage device and content storage method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To automatically generate a chapter of a scene that interests unspecified persons.SOLUTION: A content storage device 100 includes: a data storage unit 104 for storing a content; a caption analysis unit 103 for analyzing caption data included in the content to identify a display time of each caption displayed on one screen; a communication unit 113 for communicating with a network 160; an associative word acquisition unit 105 for acquiring an associative word associated with a keyword from an associative word storage unit 150 connected to the network 160; and a chapter generation unit 108 for generating chapter information indicating a chapter sectioned by the display time of the caption including the associative word.

Description

  The present invention relates to a content storage device and a content storage method.

  With the widespread use of optical disk recorders and hard disk recorders, it has become common to record video content such as television broadcasts for a long time. However, since the user's television viewing time is limited, it is desired that a desired scene can be easily found from the recorded video content.

  Therefore, a technique is known that makes it easy for a user to find an arbitrary scene by extracting a scene change due to switching of video or audio of a content and dividing it into chapters. By registering the scene change time in the video content as the chapter information, it is possible to easily skip to the start point of the desired scene.

  In addition, there is a technique for uploading metadata created by a user (description of text information such as chapter information and comments on a scene in a predetermined format) to a network and sharing it with many other users (for example, Patent Document 1). reference). With this technology, a user can generate chapters using metadata created by someone. The user can understand the contents of the scene by referring to the comment corresponding to each scene, and can easily find a desired scene.

JP 2007-110697 A

  In the technique described in Patent Document 1, it is necessary that another user creates metadata and uploads the metadata to the network for the video content that the user wants to view. Therefore, when the user views video content that is not popular (without metadata), and when there are few users of the service itself and no metadata is created, the metadata cannot be acquired.

  Therefore, an object of the present invention is to automatically generate a chapter of a scene in which an unspecified person is interested.

  A content storage device according to an aspect of the present invention includes a data storage unit that stores content, and a subtitle analysis unit that analyzes subtitle data included in the content and identifies a display time for each subtitle displayed on one screen A communication unit that communicates with a network, a related word acquisition unit that acquires a related word related to a keyword from a related word storage unit connected to the network via the communication unit, and the related A chapter generation unit that generates chapter information indicating chapters separated by display times of the subtitles including words.

  According to one embodiment of the present invention, a chapter of a scene in which an unspecified person is interested can be automatically generated.

1 is a block diagram schematically showing a configuration of a content storage device according to Embodiment 1. FIG. 3 is a flowchart showing a flow from content acquisition to chapter generation in the first embodiment. 3 is a schematic diagram illustrating a relationship between MPEG-2TS, AV data, and caption data in Embodiment 1. FIG. 6 is a schematic diagram showing subtitle management information in Embodiment 1. FIG. In Embodiment 1, it is the schematic which shows an example of the program information in television broadcasting. 6 is a schematic diagram illustrating an example of a format of related word information in Embodiment 1. FIG. 5 is a flowchart illustrating a flow of generating chapters from related words in the first embodiment. In Embodiment 1, it is the schematic for demonstrating the process at the time of producing | generating a chapter. (A) And (B) is the schematic which shows the mode of transition by the time of a related word in Embodiment 1. FIG. In Embodiment 1, it is the schematic which shows the mode of the scene search based on the produced | generated chapter information. FIG. 5 is a block diagram schematically showing a configuration of a content storage device according to a second embodiment. 10 is a flowchart showing a flow from content acquisition to chapter generation in the second embodiment.

Embodiment 1 FIG.
FIG. 1 is a block diagram schematically showing the configuration of the content storage device 100 according to the first embodiment. The content storage device 100 is connected to a network 160, and at least one related word storage unit 150 is connected to the network. The network 160 is the Internet, for example. The related word storage unit 150 is a database that aggregates search histories of unspecified users stored in a search site, for example.

  The content storage apparatus 100 includes a tuner unit 101, a recording / playback control unit 102, a caption analysis unit 103, a data storage unit 104, a related word acquisition unit 105, a chapter generation unit 108, a temporary storage unit 109, a chapter, and the like. An information storage unit 110, a remote control reception unit 111, a remote control 112, and a communication unit 113 are provided. In addition, an antenna 130 and a monitor 131 are connected to the content storage device 100. Although not shown, the antenna 130 receives television broadcast waves transmitted from a plurality of broadcast stations, for example. The monitor 131 is a display unit that displays video.

  The tuner unit 101 receives a broadcast signal corresponding to a broadcast wave selected by a user operation or the like via the antenna 130, and appropriately performs demodulation processing, compression code processing, and the like on the broadcast signal to obtain content and program information. Is generated.

The recording / playback control unit 102 separates AV data from the reception signal generated by the tuner unit 101 and stores the AV data in the data storage unit 104. The recording / playback control unit 102 controls processing for playing back the content stored in the data storage unit 104.
The caption analysis unit 103 extracts caption data from the reception signal generated by the tuner unit 101 and stores it in the data storage unit 104. At this time, the caption data is stored together with a time stamp indicating a display time for each caption displayed on one screen.
The data storage unit 104 stores content. The content includes AV data and caption data.

  The related word acquisition unit 105 acquires a related word related to the keyword from the related word storage unit 150 connected to the network 160 via the communication unit 113. Here, the related word acquisition unit 105 includes a program information management unit 106 and a program information storage unit 107.

The program information management unit 106 stores, in the program information storage unit 107, the program information included in the received signal generated by the tuner unit 101 related to the content stored in the data storage unit 104. “Program information” includes content titles, subtitles, performer information, program outlines, and the like included in EPG information of television broadcasts.
The program information management unit 106 extracts specific information as a keyword from the program information stored in the program information storage unit 107. Then, the program information management unit 106 acquires a related word related to the extracted keyword from the related word storage unit 150 connected to the network 160 via the communication unit 113. The program information management unit 106 gives the acquired related word to the chapter generation unit 108. The “related word” is a word related to the keyword. Details will be described later.
The program information storage unit 107 stores the program information given from the program information management unit 106.

  The chapter generation unit 108 extracts a part where the related word acquired by the program information management unit 106 is included in the caption data stored in the data storage unit 104. A chapter is generated based on the time stamp of the extracted part, and chapter information indicating the generated chapter is stored in the chapter information storage unit 110.

The remote control receiving unit 111 receives a radio signal from the remote control 112 and gives the content indicated by the received radio signal to the recording / playback control unit 102.
The remote controller 112 receives an instruction input from the user and transmits a radio signal indicating the content of the instruction.
Here, the remote control receiving unit 111 and the remote control 112 constitute an input unit that receives an input of an instruction from the user.
The communication unit 113 communicates with the network 160.

Next, a content reproduction method will be described.
The recording / playback control unit 102 outputs the content stored in the data storage unit 104 to the monitor 131 for playback. At this time, the recording / playback control unit 102 uses the chapter information stored in the chapter information storage unit 110 to respond to a scene search from the user. The scene search is a function of presenting the contents of the scene of each chapter to the user on the basis of the chapter information by a thumbnail or the like and reproducing the scene from the start point of the chapter selected by the user.

  Note that the content stored in the data storage unit 104 may not be obtained from television broadcasting. For example, the content may be obtained via a bus such as USB or IEEE 1394, or via a removable medium such as a DVD / BD or a memory card. Alternatively, the content may be downloaded from the network 160 via the communication unit 113. Here, in this embodiment, an MPEG-2 transport stream (TS) used for television broadcasting is assumed as content data. However, if the data is accompanied by character information such as captions, the MPEG-2 transport stream (TS) is used. -2 program stream, analog AV signal, and the like.

FIG. 2 is a flowchart showing a flow from content acquisition to chapter generation.
First, the recording / playback control unit 102 acquires AV data included in the received signal generated by the tuner unit 101 based on a broadcast signal from the outside, and the program information management unit 106 is included in this received signal. Program information is acquired (S10). The program information will be described later.

  Next, the caption analysis unit 103 extracts the caption data included in the reception signal generated by the tuner unit 101 and stores it in the data storage unit 104 (S11).

Here, in order to show how AV data and caption data are included in the received signal, a data structure of content when the received signal is MPEG-2TS will be described with reference to FIG. FIG. 3 is a schematic diagram showing the relationship between MPEG-2TS, AV data, and caption data. MPEG-2TS is configured by multiplexing PES (Packetized Elementary Stream) of video, audio, and subtitles. Each PES includes actual data of video, audio, and subtitles, and a time stamp for displaying them.
The caption data is divided in units of pages to be displayed on one screen. Using the time stamp, the display start time and the display end time can be managed for each page. The caption analysis unit 103 analyzes the caption PES separated from the received signal, associates the character information with the time stamp, creates the caption management information 140 as shown in FIG. To accumulate.

FIG. 4 is a schematic diagram showing the caption management information 140. The caption management information 140 is information in a table format having a time stamp field 140a and a caption data field 140d.
The time stamp field 140a stores the display time for displaying the caption data of the caption data per page stored in the caption data field 140d. Here, the time stamp field 140a has a display start time field 140b and a display end time field 140c.
The display start time column 140b stores a display start time for displaying subtitles of subtitle data per page stored in the subtitle data column 140d.
The display end time 140c stores the display end time for displaying the caption data of the caption data per page stored in the caption data column 140d.
The caption data column 140d stores caption data (here, text data) per page displayed in the period specified by the time stamp column 140a.

  Note that the caption analysis unit 103 may recognize characters from character information such as telop displayed in the program in addition to the caption data, and accumulate the character data of the recognized characters in the caption management information 140 as caption data. .

  Returning to the description of FIG. 2, in step S <b> 12, the program information management unit 106 extracts a guest performer name as specific information from the program information stored in the program information storage unit 107.

Here, FIG. 5 is a schematic diagram showing an example of program information 141 in television broadcasting. The program information 141 shown in FIG. 5 includes a title line 141a, a date / time line 141b, a broadcast station line 141c, a genre line 141d, a program summary line 141e, and a program detail line 141f.
The title row 141a stores the title of the program.
The date and time row 141b stores the date and time when the program is broadcast.
The broadcast station line 141c stores broadcast station information indicating a broadcast station where the program is broadcast.
The genre row 141d stores the genre of the program.
The program summary line 141e stores a summary of program contents.
The program details row 141f stores details of program contents.

  In the program information 141 shown in FIG. 5, performer information indicating the performers of the program is written in the program detail line 141f after “[Performer]” or “[Guest]”. In the present embodiment, the program information management unit 106 extracts the names of guest performers who are particularly interested in the program among the performers. Therefore, the program information management unit 106 extracts a character string following the character string “[Guest]”. In TV broadcast program information, in addition to the example of FIG. 5, the name of the performer may be written after the character string “Guest:” or “This guest is”. In addition, in the case of programs that are broadcast regularly, such as weekly and daily, series names are excluded from the names of performers, and the names that appear only in each broadcast are extracted. , Guest performer names can be extracted.

  Returning to the description of FIG. 2, in step S <b> 13, the program information management unit 106 uses the guest performer name extracted in step S <b> 12 as a keyword, and the related word storage unit 150 connected to the network 160 via the communication unit 113. Get related word from.

Here, FIG. 6 is a schematic diagram showing an example of the format of the related word information 142 stored in the related word storage unit 150. The related word information 142 is information in a table format having a rank column 142a and a related word column 142b.
The rank column 142a stores a rank in which the related words stored in the related word column 142b increase in order from the most frequently searched words together with the keyword. Here, natural numbers are assigned in order from “1” so that the higher the number, the lower the order.
The related word column 142b stores a related word that is a word searched on the search site together with the keyword.

Here, the related word information 142 is, for example, a total of words searched together with keywords on a search site on the Internet 160. In other words, the words are stored in the related word information 142 as related words in order from the word searched most frequently together with the keyword on the search site of the Internet 160.
As another example, the words may be stored in the related word information 142 as related words in order from the word that appears more frequently with the keyword on the Web page.
In a search site or the like, a service is provided that enables a more narrowed search by displaying a plurality of related words when a user inputs a keyword and allowing the user to select from them. Therefore, the program information management unit 106 can connect to the search site via the communication unit 113 and the Internet 160 and acquire a related word from the keyword. The related words are updated as needed based on data used by unspecified users on the Internet 160.
In this way, a word that is frequently searched together with a keyword or a word that frequently appears together with a keyword on a Web page is used as a related word. It can be thought of as a word indicating interest. The order of the related word information 142 can be considered to be the order in which an unspecified user shows interest.

  In addition, the keyword used for acquisition of a related word may be any information as long as it is information included in the program information 141 other than the name of the guest performer. For example, the keyword may be a program subtitle or an outline. A related word may be acquired based on a plurality of keywords.

  Returning to the description of FIG. 2, in step S <b> 14, the chapter generation unit 108 searches whether the related word acquired by the program information management unit 106 in step S <b> 13 appears in the caption data stored in the data storage unit 104. . Then, when it appears, the chapter generation unit 108 acquires the display start time corresponding to the subtitle portion from the subtitle management information 140 shown in FIG. 4, and uses the display start time as a chapter break. Information is generated and stored in the chapter information storage unit 110. The process in step S14 will be described in detail with reference to FIG.

  FIG. 7 is a flowchart showing a flow of generating chapters from related words in step S14 of FIG. FIG. 8 is a schematic diagram for explaining processing when generating a chapter.

  First, the chapter generation unit 108 sets the index n to “1” and the index k to “0” (S20). The index n is an index indicating the ranking of search words. The acquired related words are ranked as shown in FIG. 6, and are used for the search in order from the first. The index k is an index indicating the number of detected related words of the index n in the caption data at each time point.

  Next, the chapter generation unit 108 determines whether or not the number of chapters is greater than or equal to a threshold value x (S21). The chapter generation unit 108 repeats the processes of steps S22 to S30 described later until the number of chapters is equal to or greater than the threshold value x. The threshold value x is the minimum value of the final total number of chapters. The threshold value x is determined according to the content length. In a long-time content, it is considered that more scenes exist, and in order to find a desired scene, a large number of chapters are required. For example, if the content is 30 minutes, the threshold value x is “5”, and if the content is 2 hours, the threshold value x is “20”. The longer the content is, the larger the threshold value x becomes. It is desirable that the threshold value x is determined. In the example shown in FIG. 8, the value of the threshold value x is “5”. Further, the value of the threshold value x may be a constant for reducing processing, or may be set by the user according to preference.

  Next, the chapter generator 108 detects whether the nth related word appears in the caption data of the content (S22). The detection is performed from the top of the caption data in time order. When the n-th related word is found in the caption data (S22: Yes), the chapter generation unit 108 proceeds to the process of step S23, and when it is not found (S22: No), the chapter is generated. The generation unit 108 proceeds to the process of step S28.

  In step S23, the chapter generation unit 108 determines whether or not the index k is “0”. Then, when the index k is not “0” (S23: No), that is, when the nth related word has already been detected, the chapter generation unit 108 proceeds to the process of step S24, and the index k Is “0” (S23: Yes), that is, when the nth related word is detected for the first time, the process proceeds to step S25.

  In step S24, the chapter generation unit 108 starts broadcasting of the caption data that includes the nth related word detected this time and the caption data that includes the nth related word that appears immediately before. It is determined whether or not the time from the broadcast start time is within a predetermined time y. Then, when the time is longer than the time y (S24: No), the chapter generation unit 108 proceeds to the process of step S25, and when the time is within the time y (S24: Yes), the step The process proceeds to S27.

In step S25, the chapter generation unit 108 causes the temporary storage unit 109 to store the display start time of the nth related word found in step S22.
Then, the chapter generation unit 108 adds “1” to the index k, and proceeds to the process of step S27.

  In step S27, the chapter generation unit 108 updates the search position, continues the search for the nth related word, and returns to the process of step S22.

Here, the processing in steps S24 to S27 will be described with reference to FIG.
As indicated by reference numeral 143a, when a plurality of search words appear within a short period of time, it is considered that the matter continues to be talked about in that section. Therefore, although it is a continuous topic, it is inappropriate to divide chapters. Therefore, the chapter generation unit 108 does not divide chapters based on related words detected within a short time. In the example shown in FIG. 8, the value of time y is 2 minutes. The value of the time y may be set as a constant in consideration of the time to talk about one matter, but this value may be set by the user.

  Returning to the description of FIG. 7, when the nth related word is not detected from the remaining caption data in step S22 (S22: No), the chapter generation unit 108 proceeds to the process of step S28. In step S28, the chapter generation unit 108 determines whether or not the value of the index k is smaller than a threshold z that determines the maximum number of detected related words. Then, when the value is smaller than the threshold value z (S28: Yes), the chapter generation unit 108 proceeds to the process of step S29. When the value is equal to or larger than the threshold value z (S28: No), the chapter generation unit 108 proceeds to step S30. Proceed to the process.

In step S29, when the value of the index k is not “0”, the chapter generation unit 108 sets a chapter with the broadcast start time stored in the temporary storage unit 109 as a chapter as chapter information.
Then, chapter generation unit 108 adds “1” to index n, returns index k to initial value “0”, and returns to step S21. At this time, the chapter generation unit 108 deletes the broadcast start time stored in the temporary storage unit 109.

Here, the processing in steps S28 and S29 will be described with reference to FIG.
For example, as indicated by reference numeral 143b in FIG. 8, when the value of the index k is equal to or greater than the threshold value z, the related word appears many times throughout the entire program. For example, the related word can be determined not to be related to the topic of the program such as the nickname of the performer such as “Yama-chan”. For this reason, when the value of the index k is greater than or equal to the threshold value z, the chapter generation unit 108 does not perform chapter separation based on such related words. Note that the value of the threshold value z is determined according to the length of the program. For example, it is desirable to increase the value of the threshold z as the broadcast time of the program is longer. In the example shown in FIG. 8, the value of the threshold value z is “5”.

As indicated by reference numeral 143c in FIG. 8, when the cumulative number of chapters is equal to or greater than the threshold value x (here, “5”), the chapter generation unit 108 ends the flow in FIG. To do.
As described above, chapters are generated according to the flow described above.

  The timing at which the chapter is generated in the flow shown in FIG. 7 is preferably performed when the content has been recorded. Moreover, it may be performed when the user designates, or may be performed when a certain time has elapsed after the end of broadcasting.

  An advantage in the case where chapter generation is performed after a certain time has elapsed after the end of broadcasting will be described. As described above, the related word acquired by the program information management unit 106 using a keyword is updated at any time according to the usage history of an unspecified user. A certain related word becomes higher after a certain period of time by searching on a search site or posting on a website such as a blog for a topic that has been viewed by a person who has previously viewed the content There is a possibility of coming. The related word is highly likely to be a topic in the content, and can be determined to be important for many people. Therefore, by acquiring a related word after a certain time has elapsed, it is possible to generate a chapter that is more suited to the content.

FIGS. 9A and 9B are schematic diagrams showing the transition of related words over time. FIG. 9A shows the related word information 142 # 1 immediately after the broadcast, and FIG. 9B shows the related word information 142 # 2 after a predetermined time has passed since the broadcast.
The related word information 142 # 2 after the elapse of a certain time from broadcasting newly includes the word “XX beauty” that was not included in the related word information 142 # 1 immediately after broadcasting. Further, in the related word information 142 # 2 after a certain time has passed since the broadcast, the word “stage” having a lower rank in the related word information 142 # 1 immediately after the broadcast is included in the upper rank. In this way, by using the related word information 142 # 2 after a certain time from the broadcast, chapters related to the related words included in the higher order at a certain time that are not included in the related word information 142 # 1 immediately after the broadcast are generated. it can.

  Note that the related word information 142 # 1 immediately after the broadcast is also stored in the content storage device 100, so that the chapter generator 108 can compare with the related word information 142 # 2 after a certain time has passed since the broadcast. The rapidly rising word may be preferentially collated with the caption data. For example, the chapter generation unit 108 determines a word in which a value obtained by subtracting a rank after a certain time from broadcasting is larger than a predetermined threshold from a rank immediately after broadcasting earlier than the order determined by the rank of the word, It can be collated with subtitle data.

  FIG. 10 is a schematic diagram showing a scene search based on the generated chapter information. The recording / playback control unit 102 displays a scene search screen 144 on the monitor 131 in response to a request from the user when playing back content. At this time, the user can easily grasp the contents of the scene by displaying not only the thumbnail 144a of each scene but also the character information screen 144b that displays a part of the caption data including the related word.

  In addition, the user can select a related word used for chapter generation, and only a scene of interest can be reproduced.

  Through the above processing, a scene in which many people are interested in a topic can be automatically extracted even if another user does not generate a chapter in advance.

  Further, in a program such as a talk program with few image switching, it is difficult to detect a scene change by a method of automatically recognizing video and audio switching and dividing a chapter. However, in the present invention, chapters are generated based on character information based on related words and subtitle data, so that chapters can be appropriately divided according to the contents spoken by the performers.

  Further, even when it is difficult to grasp the contents of a scene only with thumbnails, the contents of the scene can be easily grasped by using character information.

Embodiment 2. FIG.
FIG. 11 is a block diagram schematically showing the configuration of the content storage device 200 according to the second embodiment. The content storage device 200 includes a tuner unit 101, a recording / playback control unit 202, a caption analysis unit 103, a data storage unit 104, a related word acquisition unit 205, a chapter generation unit 108, a temporary storage unit 109, a chapter, and the like. An information storage unit 110, a remote control reception unit 211, a remote control 212, and a communication unit 113 are provided. The content storage device 200 according to the second embodiment is different from the content storage device 100 according to the first embodiment in a recording / playback control unit 202, a remote control reception unit 211, a remote control 212, and a related word acquisition unit 205.

  The recording / playback control unit 202 obtains a reception signal including content related to free words stored in the free word storage unit 215 from the tuner unit 101 via the antenna 130, separates AV data from the reception signal, and stores the data. Stored in the unit 104. For example, the recording / playback control unit 202 acquires a reception signal of content in which the free word stored in the free word storage unit 215 is included in at least one of a title, a program overview, and program details.

The remote controller 212 performs the same processing as in the first embodiment, and accepts input from the user as a free word for matters of interest. The remote control 212 transmits a radio signal indicating the input free word.
In addition to performing the same processing as in the first embodiment, remote control receiving unit 211 receives a radio signal transmitted from remote control 212, identifies a free word input from this radio signal, and frees the identified free word. This is given to the word management unit 214.

  The related word acquisition unit 205 is connected to the related word storage unit 150 via the communication unit 113 and the network 160 to acquire a related word related to the keyword. Here, the related word acquisition unit 205 includes a free word management unit 214 and a free word storage unit 215.

The free word management unit 214 stores free words input by the user via the remote control 212 in the free word storage unit 215.
Further, the free word management unit 214 extracts the free words stored in the free word storage unit 215 as keywords. Then, the free word management unit 214 acquires a related word related to the extracted keyword from the related word storage unit 150 connected to the network 160 via the communication unit 113. The free word management unit 214 gives the acquired related word to the chapter generation unit 108.
The free word storage unit 215 stores the free word given from the free word management unit 214.

  FIG. 12 is a flowchart showing a flow from content acquisition to chapter generation in the second embodiment.

  First, the remote control 212 accepts an input from the user as a free word (S40). Then, the remote control 212 transmits a radio signal indicating the input content. The remote control receiving unit 211 receives a radio signal transmitted from the remote control 212, specifies a free word input from the received signal, and supplies the specified content to the free word management unit 214. The free word management unit 214 stores free words acquired via the remote control reception unit 211 in the free word storage unit 215.

  Next, the recording / playback control unit 202 acquires a reception signal including content related to the free word stored in the free word storage unit 215 from the tuner unit 101 via the antenna 130, and separates AV data from the reception signal. The data is stored in the data storage unit 104 (S41).

  Next, the caption analysis unit 103 extracts caption data included in the received signal acquired from the tuner unit 101, and stores it in the data storage unit 104 (S42).

  Next, the free word management unit 214 acquires a related word from the related word storage unit 150 connected to the network 160 via the communication unit 113 using the free word input in step S40 as a keyword (S43). .

  The chapter generation unit 108 searches whether the related word acquired by the free word management unit 214 in step S43 appears in the caption data stored in the data storage unit 104. Then, when it appears, the chapter generation unit 108 acquires the display start time corresponding to the subtitle portion from the subtitle management information 140 shown in FIG. 4, and uses the display start time as a chapter break. Information is generated and stored in the chapter information storage unit 110 (S44). The process in step S44 is the same as the process shown in the flowchart of FIG.

  Here, the processing from step S41 to step S44 is repeated by the number of contents if there are a plurality of contents related to the free word input in step S40.

  Also, the content storage apparatus 200 according to Embodiment 2 may generate a chapter by setting a free word input by the user as a keyword for specific content. In that case, the processing shown in FIG. 12 may be executed on the content stored in advance, excluding step S41 of FIG.

  When a plurality of keywords are extracted from program information, it is difficult to select a keyword that the user is interested in. In addition, when a matter that the user is interested in is not included in the program information, an appropriate keyword cannot be extracted. In the second embodiment, even in such a case, a chapter of a scene desired by the user can be generated.

  The content storage devices 100 and 200 according to the first and second embodiments described above may be configured as a digital broadcast reception device or a digital broadcast reception recording / playback device. In addition, content storage apparatuses 100 and 200 according to Embodiments 1 and 2 may be configured as a personal computer or a mobile terminal.

  100, 200: Content storage device 101: Tuner unit 102, 202: Recording / playback control unit 103: Subtitle analysis unit 104: Data storage unit 105, 205: Related word acquisition unit 106: Program information management unit 107: Program information storage unit, 108: Chapter generation unit, 109: Temporary storage unit, 110: Chapter information storage unit, 111, 211: Remote control reception unit, 112, 212: Remote control, 113: Communication unit, 214: Free word management Part, 215: a free word storage part.

Claims (12)

A data storage unit for storing content;
Analyzing the caption data included in the content, and specifying a display time for each caption displayed on one screen;
A communication unit for communicating with the network;
A related word acquisition unit that acquires a related word related to a keyword from a related word storage unit connected to the network via the communication unit;
And a chapter generation unit that generates chapter information indicating chapters separated by display times of the subtitles including the related words. The related word acquisition unit
A program information management unit for acquiring program information accompanying the content;
A program information storage unit that stores program information acquired by the program information management unit,
The content storage apparatus according to claim 1, wherein the program information management unit acquires the related word using specific information included in the program information as the keyword. A tuner unit for generating a reception signal including the content and the program information from a broadcast signal received via an antenna;
The content storage device according to claim 2, wherein the program information management unit acquires the related word after a predetermined time has elapsed since the tuner unit received a broadcast signal. An input unit for receiving instructions,
4. A recording / playback control unit that plays back a scene corresponding to a chapter specified by an instruction received by the input unit based on the chapter information. 5. The content storage device according to claim 1. The recording / playback control unit generates a chapter selection screen that displays at least a part of the subtitles including the related word, and receives an instruction input from the input unit on the chapter selection screen. 4. The content storage device according to 4. It further includes an input unit that receives input of instructions,
The related word acquisition unit
A free word management unit that acquires a free word indicated by an instruction that the input unit has accepted an input;
A free word storage unit for storing free words acquired by the free word management unit,
The content storage device according to claim 1, wherein the free word management unit acquires the related word using the free word as a keyword. A data storage process for storing content;
Analyzing the subtitle data included in the content and identifying a display time for each subtitle displayed on one screen;
A related word acquisition process for acquiring a related word related to a keyword from a related word storage unit connected to the network;
A chapter generation process for generating chapter information indicating chapters separated by display times of the subtitles including the related words. The related word acquisition process includes:
A program information management process for acquiring program information associated with the content;
A program information storage process for storing program information acquired in the program information management process;
The content storage method according to claim 7, further comprising: an acquisition step of acquiring the related word using the specific information included in the program information stored in the program information storage step as the keyword. . A reception process for generating a reception signal including the content and the program information from a broadcast signal received via an antenna;
The content accumulating method according to claim 8, wherein the obtaining step obtains the related word after a predetermined time has elapsed since the broadcast signal was received in the receiving step. An input process for receiving instructions,
The recording / playback control step of playing back a scene corresponding to the chapter specified by the instruction received in the input step based on the chapter information. The content storage method according to claim 1. The recording / playback control process generates a chapter selection screen that displays at least a part of the subtitles including the related word, and receives an instruction input in the input process on the chapter selection screen. 10. The content storage method according to 10. An input process for receiving an instruction input;
The related word acquisition process includes:
A free word management process for obtaining a free word indicated by an instruction that has received an input in the input process;
A free word storage process for storing free words acquired in the free word management process;
The content storage method according to claim 7, further comprising: an acquisition process of acquiring the related word using the free word stored in the free word storage process as a keyword.

Images