JP2023170822A - Explanation voice production device and program - Google Patents

Abstract

To provide an explanation voice with high expandability and versatility at a real time while using data of a plurality of information sources.SOLUTION: An analysis part 11 of an explanation voice production device 1 extracts a text element by analyzing data input from a plurality of information sources 2, and a storage part 12 stores the data to an information management table 13 by applying a label to the text element. A reading part 16 reads out the updated text element from the information management table 13 in each text element of the label defined to a template 14. A format conversion part 17 converts the label and the text element into a Json file in each speech of the read text element. A text generation part 18 generates a text for an explanation voice from the Json file. An order disposal control part 19 determines an order of the speech on the basis of a priority of a presentation timing contained in the level of the Json file, and disposes the Json file of the speech when an output processing related to the speech in ahead is completed.SELECTED DRAWING: Figure 2

Description

The present invention relates to an explanatory audio production device and program that generate text for explanatory audio.

2. Description of the Related Art Conventionally, commentary audio services have been known that broadcast sports broadcast programs and provide viewers with commentary audio of the broadcast programs (for example, see Patent Document 1).

FIG. 15 is a diagram illustrating an overview of a system that provides an audio commentary service. This system includes a broadcast transmitting device 101, a broadcast receiving device 102, an explanatory audio production and distribution device 103, an application server 104, and a mobile terminal 105.

The broadcast transmitting device 101, the commentary audio production and distribution device 103, and the application server 104 are installed, for example, at a broadcasting station, and the broadcast receiving device 102 is installed, for example, at the home of the viewer 100. Furthermore, the mobile terminal 105 is used by a viewer 100 who views a broadcast program at home.

With the commentary audio service of this system, the viewer 100 can receive commentary audio as well as an audio and video broadcast program that explains the match situation using the announcer's commentary and the commentator's commentary.

Broadcast transmitting device 101 transmits broadcast program content to broadcast receiving device 102 via digital terrestrial broadcast waves. The broadcast receiving device 102 is, for example, a television receiver, and receives the broadcast program content transmitted from the broadcast transmitting device 101 via digital terrestrial broadcast waves, and reproduces the received broadcast program content.

The commentary audio production and distribution device 103 produces commentary audio for the broadcast program content being transmitted by the broadcast transmitting device 101, and transmits the commentary audio to the mobile terminal 105. The application server 104 stores an application that operates on the mobile terminal 105, and transmits the application to the mobile terminal 105 in response to a request from the mobile terminal 105. App is an abbreviation for application, and here it is a program that receives and plays explanatory audio.

The mobile terminal 105 is, for example, a smartphone or a PDA (Personal Digital Assistant), and plays explanatory audio of the broadcast program content in synchronization with the broadcast program content received by the broadcast receiving device 102. When reproducing the commentary audio, the mobile terminal 105 changes the reproduction speed etc. according to the operation of the viewer 100.

For example, if the broadcast program is a baseball broadcast, the viewer 100 can receive video and audio of the baseball game as well as commentary audio that explains the game situation at that time in detail, so that the viewer 100 can understand the contents of the game in detail. can do. Baseball commentary audio includes, for example, pitcher information, pitcher's movements, pitch type, ball speed, course, batter's information, batter's movements, scores, etc. according to the game situation.

An example of the commentary audio production and distribution device 103 that realizes such commentary audio service is one that receives data in accordance with ODF (Olympic Data Feed) specifications, produces and distributes commentary audio using the data. A system is known (see, for example, Non-Patent Document 1).

The commentary audio production and distribution device 103 described in this non-patent document 1 sequentially receives data such as scores and fouls in the current match situation from one information source that provides Olympic data. Then, the commentary audio production and distribution device 103 generates the text of the commentary text according to the match situation by applying variables to a preset template, converts the text into audio using a speech synthesizer, and generates commentary audio. The audio file is sent to the mobile terminal 105.

JP2017-203827A

Tadashi Kumano, “Audio guide generation technology for explaining sports programs”, NHK Giken R&D, No. 154, pp. 12-20, 2017

As mentioned above, the technology in Non-Patent Document 1 can be used only at specific Olympic Games, and it generates text using a limited format of data in accordance with the ODF specifications of the Games, and generates audio files of explanatory audio. It is something that generates.

For this reason, the technology of Non-Patent Document 1 cannot be used as is for other tournaments, and has the problem of low expandability and versatility.

Furthermore, in the technology of Non-Patent Document 1, there is only one information source for generating explanatory audio, so even if there is information that you want to convey to the viewer 100 as an explanatory audio, it is not necessarily conveyed to that information source. There is no guarantee that the information you want exists. For this reason, a technology that can utilize multiple information sources has been desired.

Furthermore, with the technique disclosed in Non-Patent Document 1, there is no guarantee that data will be distributed from the information source at the required timing. For this reason, with respect to data that requires real-time performance, there is a problem in that an audio commentary service cannot be established unless the data is distributed from the information source at the timing when it should be conveyed to the viewer 100.

In this way, the technology of Non-Patent Document 1, which uses data distributed from one information source to generate text and generate explanatory audio, is insufficient as an explanatory audio service, and it is difficult to meet the demands of the viewer 100. cannot be fully satisfied.

Therefore, the present invention has been made to solve the above problems, and its purpose is to provide an explanatory voice that utilizes data from multiple information sources and that can provide extensible and versatile explanatory audio in real time. Its purpose is to provide production equipment and programs.

In order to solve the above problem, an explanation audio production device according to claim 1 is an explanation audio production device that generates a text for an explanation audio of a sports program being live distributed for each utterance. is composed of one or more text elements, a template in which utterance definition data including one or more labels corresponding to the one or more text elements is defined, and an information management table in which the text elements are stored. and inputting data corresponding to the game situation of the sports program from each of a plurality of information sources, and extracting the text element from the data by analyzing the data according to a preset data format of the information source. an analysis unit that assigns a label including the priority of the timing of presenting the utterance of the text element to the text element extracted by the analysis unit, and stores the text element to which the label has been assigned in the information management table; and an update monitor that monitors whether the text element stored in the information management table has been updated and outputs the label given to the text element when it is determined that the text element has been updated. and the utterance of the utterance definition data including the label output by the update monitoring unit, from the information management table, the corresponding one or more labels to which one or more labels included in the utterance definition data are attached a reading unit that reads out a plurality of text elements and outputs one or more labels of the utterance and one or more text elements corresponding thereto; and one or more labels of the utterance output by the reading unit and the same. a format conversion unit that format-converts one or more text elements corresponding to a file including a predetermined playback time; and extracting the one or more text elements from the file format-converted by the format conversion unit; A text generation unit that generates and outputs the text and a file for each utterance whose format has been converted by the format conversion unit are input, and based on the priority included in the label of the file for each utterance, determine the order of utterances, reset the playback time included in the file for each utterance according to the order, output the playback time included in the file of the first utterance for which the order has been determined, The present invention is characterized by comprising an order discard control unit that discards a file of utterances.

Further, in the commentary audio production device according to claim 2, in the commentary audio production device according to claim 1, the plurality of information sources include an information source that transmits real-time data according to the match situation of the sports program. and an information source that transmits data of the sports program according to an input operation by an operator, an information source that transmits data obtained by analyzing an image of a game situation of the sports program, and a game situation of the sports program. The information source includes at least one information source that transmits data obtained by recognizing the voice of the user.

Further, in the explanatory audio production device according to claim 1, the template includes, in addition to the one or more labels, the one or more labels for each utterance. One of the text elements is defined as a trigger label, and the update monitoring unit monitors whether or not the text element to which the trigger label is attached has been updated in the information management table, and it is determined that the text element has been updated. The trigger label is output.

Further, in the explanatory audio production apparatus according to claim 1, the label includes the type of the information source, the competition type of the sports program, the priority, and the text element to which the text element belongs. The group shall be composed of a group and each numerical value indicating an item within the group, and the above-mentioned information is used when reading out the corresponding one or more text elements to which the one or more labels included in the utterance definition data are attached. A label is set as a read target label, and in addition to the read target label, a group to which the text elements constituting the read target label belong and a label in which the items in the group are the same but the type of the information source is different is classified as a homogeneous label. If the information management table stores a plurality of text elements to which the same kind of label is attached, the reading unit selects the most text element from the plurality of text elements to which the same kind of label is attached from the information management table. The method is characterized in that the previously stored text element is read out.

Further, in the explanatory audio production device according to claim 5, in the explanatory audio production device according to claim 4, the reading unit stores a text element to which the reading target label is attached in the information management table. First, if a text element to which the same type label other than the read target label is stored is stored, the text element to which the same type label other than the read target label is attached is read from the information management table. Features.

Further, in the explanatory audio production device according to claim 6, in the explanatory audio production device according to claim 1, the label includes a competition event of the sports program in addition to the priority level, and the reading section The text element read from the information management table is modified according to the sport event included in the label assigned to the text element, and one or more labels and one or more labels of the utterance are added. The text element (or the modified text element, if any) is output.

Further, in the explanatory audio production device according to claim 7, in the explanatory audio production device according to claim 1, the one or more labels included in the utterance definition data include a label corresponding to a predetermined particle or word. The reading unit reads one or more text elements including the predetermined particle or word from the information management table, and the text generation unit reads the predetermined particle or word from the file. The present invention is characterized in that one or more text elements containing the predetermined particle or word are extracted, and a text containing the predetermined particle or word is generated and output.

Further, in the explanatory audio production device according to claim 8, in the explanatory audio production device according to claim 1, the priority included in the label is any one of information indicating immediate, semi-immediate, regular, and others. , the immediate priority is the highest, the semi-immediate is the next highest, and the other is the lowest, and the order discard control unit transmits the utterance of the label including the periodic priority to the a first utterance of a label that includes an immediate or said semi-immediate priority, and a second utterance of a label that includes said regular priority that is arranged at a predetermined time interval; The order of the utterances is determined so that the higher the frequency, the closer the utterances are to the beginning, and the second utterance is arranged at the predetermined time interval after the first utterance. It is characterized by determining the order of.

Further, in the explanatory audio production device according to claim 9, in the explanatory audio production device according to claim 1, the label includes the type of the information source, the competition type of the sports program, the priority, and the text element to which the text element belongs. The label shall be composed of a group and a numerical value indicating an item within the group, and the label shall be used when reading out the corresponding one or more text elements to which the one or more labels included in the utterance definition data are attached. As a read target label, in addition to the read target label, a label to which the group to which the text elements constituting the read target label belong and the items in the group are the same, and the type of the information source is different is defined as a homogeneous label; When the order discard control unit inputs a new file including the same kind of label in accordance with the update determination by the update monitoring unit and determines the order of utterances for the files for each utterance, the same kind label If there are a plurality of files including the same kind of label, the files other than the new file among the plurality of files including the same kind of label are discarded.

In the explanatory audio production device according to claim 10, in the explanatory audio production device according to claim 1, the order discard control unit discards files for which a preset time has elapsed among the files for each utterance. It is characterized by doing.

Further, the program according to claim 11 is directed to a computer forming a commentary audio production device that generates text for commentary audio of a sports program being live-distributed for each utterance. A template in which utterance definition data including one or more labels corresponding to the one or more text elements is defined when composed of text elements, an information management table in which the text elements are stored, and a plurality of information sources. an analysis unit that extracts the text element from the data by inputting data corresponding to the game situation of the sports program from each and analyzing the data according to a preset data format of the information source; a storage unit that assigns a label including a priority of the timing of presentation of the utterance of the text element to the extracted text element, and stores the text element to which the label has been assigned in the information management table; an update monitoring unit that monitors whether or not the text element stored in a management table has been updated, and outputs the label given to the text element when it is determined that the text element has been updated; Regarding the utterance of the utterance definition data that includes the label, one or more corresponding text elements to which one or more labels included in the utterance definition data are attached are read from the information management table, and a reading unit that outputs one or more labels and one or more text elements corresponding thereto; one or more labels of the utterance output by the reading unit and one or more text elements corresponding thereto; a format conversion unit that converts the format into a file including a predetermined playback time; a text generation unit that extracts the one or more text elements from the file whose format has been converted by the format conversion unit, generates and outputs the text; and inputting the file for each utterance whose format has been converted by the format conversion unit, determining the order of the utterances based on the priority included in the label of the file for each utterance, and converting the utterances according to the order. As an order discard control unit that resets the playback time included in a file for each utterance, outputs the playback time included in the file of the first utterance for which the order has been determined, and discards the file of the first utterance. It is characterized by making it function.

As described above, according to the present invention, it is possible to utilize data from a plurality of information sources and provide explanatory audio with high expandability and versatility in real time.

1 is a schematic diagram illustrating an example of the overall configuration of an explanatory audio production and distribution system including an explanatory audio production device according to an embodiment of the present invention. 1 is a block diagram showing a configuration example of an explanatory audio production device according to an embodiment of the present invention. FIG. 7 is a flowchart illustrating an example of processing by an analysis unit and a storage unit. It is a figure explaining a label. It is a figure which shows the example of a template. FIG. 3 is a diagram illustrating an example of utterance definition data defined in a template, labels and text elements stored in an information management table, a Json file, and explanatory audio text. 3 is a flowchart illustrating an example of processing by an update monitoring unit. FIG. 3 is a diagram illustrating an example of processing by an update monitoring unit. 3 is a flowchart illustrating an example of processing by a reading unit. FIG. 7 is a diagram illustrating an example of reading out a text element (step S905). 3 is a flowchart illustrating an example of processing by a format converter. 3 is a flowchart illustrating an example of processing by a text generation unit. 3 is a flowchart illustrating an example of processing by an order discard control unit. FIG. 6 is a diagram illustrating the transition of utterance data in an array in an order discard control unit. FIG. 1 is a diagram illustrating an overview of a system that provides an audio commentary service.

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments for carrying out the present invention will be described in detail using the drawings.
[Explanatory audio production and distribution system]
First, the explanation audio production and distribution system that realizes the explanation audio service will be explained. FIG. 1 is a schematic diagram illustrating an example of the overall configuration of an explanatory audio production and distribution system including an explanatory audio production device according to an embodiment of the present invention.

This explanatory audio production and distribution system 10 includes an explanatory audio production device 1, a plurality of information sources 2, a speech synthesis device 3, a distribution device 4, and a mobile terminal 5. The commentary audio production and distribution system 10 corresponds to the commentary audio production and distribution device 103 and the mobile terminal 105 among the systems that provide the commentary audio service shown in FIG.

The commentary audio production device 1 is a device that generates commentary audio text for each utterance when producing commentary audio for a sports program that is being distributed live. An explanatory audio production device 1 receives real-time data from a plurality of information sources 2 according to the match situation of a live sports program. Then, the explanatory audio production device 1 extracts text elements by analyzing the data according to the unique data format of the information source 2 that is the input source of the data, assigns a label to the text element, and stores it in the information management table 13 described later. Store.

Here, the text element is one or more elements that constitute the explanatory voice text (text of the content desired to be uttered) to be generated. A label is information for identifying the content of a text element. Details will be described later.

In order to generate explanatory audio text for one utterance, the explanatory audio production device 1 reads updated text elements from the information management table 13 and includes the playback time according to utterance definition data defined in a template 14 described later. A Json file is generated, an utterance ID is assigned, and an explanatory voice text is generated. Further, the commentary audio production device 1 determines the order of utterances and resets the playback time. The reproduction time is the time at which the mobile terminal 5 reproduces the audio file of the explanatory audio text.

For each utterance, the explanatory audio production device 1 outputs the utterance ID and explanatory audio text to the speech synthesis device 3, and also outputs the utterance ID and playback time to the distribution device 4.

The information source 2 includes a plurality of information sources for each sport event, for example. As shown in FIG. 1, the plurality of baseball information sources 2 include, for example, an information source 2-1 that distributes Olympic-related data according to ODF specifications, and an information source 2-1 that distributes professional baseball-related data according to BIS specifications. There are an information source 2-2, an information source 2-3 that distributes data related to professional baseball according to BIP specifications, and an information source 2-4 that distributes data related to high school baseball according to SIGN specifications.

In addition, the baseball information source 2 includes an information source 2-5 that transmits baseball-related data according to predetermined specifications based on input operations by an operator who is watching a broadcast program, and an information source 2-5 that analyzes images of baseball game situations. Information source 2-6 generates baseball-related data by doing so and transmits baseball-related data according to predetermined specifications; There are information sources 2-7, etc. that transmit baseball-related data according to the specifications. In addition, there are multiple information sources 2 that distribute tennis data.

In this way, in addition to the information sources 2-1, ..., 2-4 that transmit real-time data according to the match situation of sports programs, the information source 2-5 based on operator input operations, and the information source based on image analysis. 2-6 and information source 2-7 by voice recognition, or by using at least one of information sources 2-5, 2-6, 2-7, information sources 2-1, ..., 2- It is possible to supplement the missing data distributed from 4, and it is possible to widen the range of explanatory voices to be presented.

Further, by using data from a plurality of information sources 2, it is possible to realize a highly versatile commentary audio production and distribution system 10 without depending on a specific program, tournament, or competition. Furthermore, since the data necessary for the explanatory audio can be acquired from a plurality of information sources 2, it becomes possible to present the explanatory audio reliably, and it is also possible to realize real-time performance.

The speech synthesis device 3 receives the utterance ID and the commentary audio text from the commentary audio production device 1, and generates a synthesized sound from the commentary audio text using existing technology to generate an audio file. Then, the speech synthesis device 3 outputs the utterance ID and the audio file to the distribution device 4.

The distribution device 4 inputs the utterance ID and playback time from the commentary audio production device 1, and also inputs the utterance ID and audio file from the speech synthesis device 3. Then, the distribution device 4 changes the playback time so that the utterance of the commentary audio does not overlap with the utterance of the main audio of the broadcast. Specifically, the distribution device 4 predicts the end time of the utterance of the main audio of the broadcast, and predicts the time period from the start time of the main audio to the end time and the time during which the audio file of the explanatory audio will be played. If the periods overlap, the playback time is changed to a time after the end time of the main audio.

The distribution device 4 distributes the audio file and playback time of the same utterance ID to the mobile terminal 5.

The mobile terminal 5 receives the audio file and playback time distributed from the distribution device 4, and plays the audio file at the playback time.

For audio that progresses in real time, such as in a live broadcast, there may be situations in which multiple audio overlap is unavoidable. Furthermore, elderly people or people with hearing loss may have difficulty hearing women's voices, or may miss hearing voices in fast-paced competitions. In this case, the mobile terminal 5 selects the speaker and reproduction speed of the reproduced audio according to the operation of the viewer 100, and reproduces the audio. This makes it possible to achieve ease of listening according to each situation.

[Explanatory audio production device 1]
Next, the explanation audio production device 1 shown in FIG. 1 will be explained in detail. FIG. 2 is a block diagram showing a configuration example of the explanatory audio production device 1 according to the embodiment of the present invention.

This explanatory audio production device 1 includes an analysis section 11, a storage section 12, an information management table 13, a template 14, an update monitoring section 15, a reading section 16, a format conversion section 17, a text generation section 18, and an order discard control section 19. ing.

<Analysis section 11>
FIG. 3 is a flowchart showing a processing example of the analysis unit 11 and storage unit 12. The analysis unit 11 inputs data corresponding to the match situation of the live sports program from the plurality of information sources 2 (step S301). The input data is data defined in various formats such as fixed length, CSV, XML, and Json.

The analysis unit 11 identifies the type of the information source 2 that is the data input source and generates identification information, and also analyzes the input data of the information source 2 according to a preset data format of the information source 2. Then, text elements are extracted from the data (step S302). Furthermore, when extracting a text element, the analysis unit 11 generates an analysis result indicating the type, content, etc. of the text element, and stores the text element, analysis result, and identification information in the storage unit 12. Output to.

For example, the analysis unit 11 inputs professional baseball related data ("Pitcher Suzuki", "Kamaeta", etc.) according to the BIS specifications from the information source 2-2, and determines that the type of the information source 2 is "BIS" as identification information. ” is generated and analyzed in accordance with the data format of information source 2-2, thereby extracting the text elements “Pitcher Suzuki” and “Kamaeta” from the data. Furthermore, the analysis unit 11 generates results indicating that the sport event is "baseball," "Pitcher Suzuki" is the name of the pitcher, and "kamaeta" is the motion of the pitcher.

<Storage section 12>
The storage unit 12 inputs the text element, the analysis result, and the identification information from the analysis unit 11, and assigns a label to the text element according to the analysis result and identification information (step S303). Then, the storage unit 12 stores the labeled text element together with the time stamp in the information management table 13 (arranges it at a position according to the label) (step S304). The timestamp is information regarding the time when the text element is stored in the information management table 13.

As mentioned above, a label is information for identifying the content of a text element. As shown in FIG. 4, which will be described later, the label is composed of a total of five numerical values from the first column to the fifth column. Specifically, the first column shows the type of information source 2 from which the text element was obtained, the second column shows the competition type of the text element, and the third column shows the presentation timing priority when presenting the text element as an utterance ( priority). Furthermore, the fourth and fifth columns show groups and items when text elements are classified according to content.

For example, the storage unit 12 receives the text element “Pitcher Suzuki” from the analysis unit 11, analysis results (results indicating that the sport is “baseball” and “Pitcher Suzuki” is the name of the pitcher, etc.), and identification information (information (information indicating that the type of source 2 is "BIS"), "2-1-3-9-1" is added as a label for the text element "Pitcher Suzuki" according to the analysis result and identification information. Grant.

As shown in FIG. 4, which will be described later, "2" in the first column indicates that the type of information source 2 is "BIS", "1" in the second column indicates that the sport is "baseball", and "3" indicates that the type of information source 2 is "BIS". Column "3" indicates that the presentation timing is "regular". Further, "9-1" in the fourth and fifth columns indicates that the group is "pitcher information" and the item is "name" (indicating that it is the name of the pitcher).

In addition, the storage unit 12 receives the text element “Kamaeta” from the analysis unit 11, and the analysis result (information source 2 is “BIS”, the sport event is “baseball”, and “Kamaeta” is the motion of the pitcher). etc.) and identification information (information indicating that the type of information source 2 is "BIS"), the text element "Kamaeta" will be labeled with "," according to the analysis result and identification information. 2-1-1-11-1”.

As shown in FIG. 4, which will be described later, "2" in the first column indicates that the type of information source 2 is "BIS", "1" in the second column indicates that the sport is "baseball", and "3" indicates that the type of information source 2 is "BIS". Column "1" indicates that the presentation timing is "immediate". Further, "11-1" in the fourth and fifth columns indicates that the group is "pitcher's action" and the item is "kameta."

As a result, a common label is given to the text elements, so it is possible to centrally manage the text elements extracted from the data in the data format of the information source 2, which is originally different, in the information management table 13. can. Similarly, data for different sports can be centrally managed.

Here, in the storage unit 12, the first column of the label is given a numerical value according to the identification information, the second, fourth, and fifth columns of the label are given numerical values according to the analysis results, and the third column The presentation timing of the eyes is given a numerical value in the third column of the label defined in the template 14, which will be described later. Specifically, the storage unit 12 first determines the numerical values in the 1st, 2nd, 4th, and 5th columns according to the analysis result and identification information regarding the numerical values in the 1st to 5th columns of the label to be attached. Then, for the third column, the storage unit 12 stores the values in the 1st, 2nd, 4th, and 5th columns that are the same as the determined values in the 1st, 2nd, 4th, and 5th columns among the labels defined in the template 14 to be described later. The label having the label is specified, the numerical value in the third column of the specified label is extracted, and the numerical value is determined as the numerical value in the third column of the label to be provided.

Note that the presentation timing of the third column may be defined in the template 14, which will be described later, in accordance with the fourth and fifth columns of the labels. In this case, the template 14 includes a numerical value of the presentation timing of the third column for each of the fourth and fifth columns of the label, and the storage unit 12 stores the numerical value of the presentation timing of the third column in each of the fourth and fifth columns of the label. , After determining the numerical value in the fifth column, read out the numerical value of the presentation timing in the third column corresponding to the fourth and fifth columns of the label from the template 14, and use the read numerical value as the numerical value in the third column of the label to be given. decide.

<Information management table 13>
The information management table 13 stores labeled text elements together with time stamps. That is, the information management table 13 is composed of labels, text elements, and time stamps. The text element is the minimum unit element when configuring the explanatory audio text.

FIG. 4 is a diagram illustrating labels. As shown in FIG. 4(1), a label is given to one text element and is composed of a total of five numerical values from the first column to the fifth column.

As shown in FIG. 4(2), the first column of labels indicates the type of information source 2, which is the distribution source of the text element. The type of information source 2 is information for distinguishing features. The numerical value "1" indicates "ODF", the numerical value "2" indicates "BIS", the numerical value "3" indicates "BIP", the numerical value 4 indicates "image analysis tool", the numerical value 5 indicates "input tool", and so on.

The second column of labels indicates the competition event expressed by the content of the text element, as shown in FIG. 4(3). The competition event is information that is used when different unique utterances are required for each competition, or when different conditions are required for each competition. The number "1" indicates "baseball", the number "2" indicates "tennis", the number "3" indicates "table tennis", the number "4" indicates "badminton", the number "5" indicates "basketball", etc. .

The third column of labels indicates the priority of presentation timing when presenting a text element as an utterance, as shown in FIG. 4(4). The presentation timing is information for controlling the timing of presenting the explanatory audio. The numerical value "1" indicates "immediate," the numerical value "2" indicates "semi-immediate," the numerical value "3" indicates "regular," and the numerical value "4" indicates "other."

It is necessary to present explanatory audio one by one, and the presentation timing is set in advance according to the text element, especially under the condition that the explanatory audio is presented in conjunction with the broadcast, from the viewpoint of whether or not the broadcast may overlap with the explanatory audio. be done.

"Immediately" means that it is important to synchronize with the video, and the application on the mobile terminal 5 plays it immediately as soon as the audio file of the commentary audio is distributed from the distribution device 4, without considering any overlap with the broadcast audio. do. Therefore, it is placed at the highest priority level. For example, if the explanatory audio is ``The pitcher caught'' or ``The pitcher threw,'' these words are meaningless unless they are played in synchronization with the video.

"Semi-immediate" means that the app on the mobile terminal 5 plays the audio file of the commentary audio within a predetermined time, taking into consideration the overlap with the broadcast audio. For example, when a technique is decided in a table tennis match, and the explanatory audio is "Suzuki vs. Yamada 10 vs. 6," the app on the mobile device 5 will set a time limit of 2 seconds, for example, to allow the player to speak without overlapping with the broadcast audio. Playback is performed when there is no overlap with the broadcast audio within 2 seconds, or playback is performed immediately when the time exceeds 2 seconds.

"Regularly" is set when the explanatory voice is not immediate and should be uttered periodically, such as the match title, match card, current score information, etc. The application on the mobile terminal 5 reproduces the audio file of the explanatory audio at predetermined time intervals or under predetermined conditions.

In this case, for example, the distribution device 4 shown in FIG. 1 receives the utterance ID and playback time as well as the label corresponding to the utterance from the commentary audio production device 1, and determines that the label presentation timing is "regular". Then, the playback time is changed so that the utterance of the commentary audio does not overlap with the utterance of the main audio of the broadcast.

The fourth column of labels indicates groups when text elements are classified according to content, as shown in FIG. 4(5). A group is information indicating a category of text elements. The number "1" is "match information", the number "2" is "game type", the number "9" is "pitcher information", the number "10" is "batter information", the number "11" is Indicates the "pitcher's action"...

As a result, the text elements are managed in groups, so that the information in the fifth column defined at the lower level can be controlled all at once.

The fifth column of labels indicates the items within the group, as shown in FIG. 4(5). Items are information obtained by further classifying categories of text elements, and are the most concrete information.

For example, if the group in the fourth column of the label is "match information" with the number "1", the number "1" in the item is "tournament name", the number "2" is "match name (for example, X vs. Y)", The numerical value "3" indicates "venue (for example, Z Stadium)", . . . For example, if the group in the fourth column of the label is "pitcher information" with the number "9", the number "1" in the item is "name (e.g. Suzuki)" and the number "2" is "season performance (e.g. current The number "3" indicates "today's performance (for example, today's ERA 0.50)". For example, if the group in the fourth column of the label is "Pitcher's action" with the number "11", the item number "1" is "Kate", the number "2" is "Throw", and the number "3" is "Pitcher's action". ” indicates “check”.

In "Group" in the fourth column of the label, the numerical values "1" to "5" and "18" are information common to all competitions, and are not used if this type of text element cannot be obtained from the information source 2. The numbers "6", "7", "15" to "17" in "Group" in the fourth column of the label are common information for racket sports, for example, if the "competition" is "table tennis", "badminton", Used in the case of "tennis". When the "competitive event" is "table tennis," "badminton," and "tennis," there are many "items" in common, so such common information is used.

The numbers "8" to "14" in "Group" in the fourth column of the label are information when "Sports Event" is "Baseball", but they are also common when "Sports Event" is "Softball". Since there is an "item", the information may be common to "baseball" and "softball".

In this way, by using multiple information sources 2, text elements can be stored in many "items" of the information management table 13, so many types of explanatory audio texts can be generated. You can expand the range of explanatory voices you want to express.

Also, if the group in the fourth column of the label is a "particle" with the number "40", the number "1" in the item is "ha", the number "2" is "no", and the number "3" is "he". , the numerical value "4" indicates "ga". Also, if the group in the fourth column of the label is "word (position)" with the number "41", the number "1" in the item is "toward", the number "2" is "toward", and the number "3" is "towards". ” indicates “forward”.

If the group in the fourth column of the label is a "particle" with the numerical value "40", and if the group in the fourth column of the label is "word (position)" with the numerical value "41", these text elements are Rather than being acquired from the information source 2, it is stored in advance in the information management table 13 as a fixed character string.

By using text elements such as "particles" or "words (positions)," that is, by using fixed text elements that are not acquired from the information source 2 and stored in advance in the information management table 13, flexibility can be achieved. It is possible to generate explanatory audio text for expressions. The application on the mobile terminal 5 can play an audio file of an explanatory voice that is similar to human speech, and the viewer 100 can easily recognize the explanatory voice.

<Template 14>
FIG. 5 is a diagram showing an example of the template 14. In this template 14, utterance definition data consisting of an utterance number, utterance content, label combination, and trigger label is defined for each explanatory voice text generated by the explanatory voice production device 1, that is, for each utterance. In order to newly increase the number of types of explanatory sounds, it is sufficient to add utterance definition data of new utterances, that is, utterance numbers, utterance contents, label combinations, and trigger labels, to this template 14.

The utterance definition data defined in the template 14 is set by key input by the user who operates the explanatory audio production device 1. Note that the configuration of the template 14 shown in FIG. 5 is an example, and other configurations may be used.

The utterance number is a number for identifying utterance definition data for each utterance. The utterance content is the content to be uttered, and is composed of one or more "items" of text elements ("items" in the fifth column of labels). A combination of labels is made up of one or more labels corresponding to the content of the utterance. The trigger label is a label corresponding to a text element whose update is monitored by the update monitoring unit 15, which will be described later.

In the example of FIG. 5, for utterance number 1, the utterance contents are "Pitcher information name" and "Pitcher's action (kamaeta)", and the label combinations are "4-1-3-9-1" and "5- 1-1-11-1" and the trigger label is "5-1-1-11-1."

This means that when the text element in the 4th and 5th columns "11-1" of the trigger label "5-1-1-11-1" stored in the information management table 13 is updated, the information management table 13 "9-1" in the 4th and 5th columns of the stored label "4-1-3-9-1" and "11-1" in the 4th and 5th columns of "5-1-1-11-1" This indicates that an explanatory audio text consisting of the text elements ``Pitcher information name'' and ``Pitcher's motion (kamaeta)'' is generated.

In addition, as utterance number 2, the utterance content is "tournament name", "match name", "country name 1", and "country name 2", and the label combination and trigger label are "1-1-3-1-1" and "1-1". -3-1-2,” “1-1-3-3-1,” and “1-1-3-3-2” are defined.

This is "1-1" in the 4th and 5th columns of the trigger label "1-1-3-1-1" stored in the information management table 13, 4 in the "1-1-3-1-2", 5th column "1-2", 4th and 5th columns "3-1" of "1-1-3-3-1" and 4th and 5th columns "1-1-3-3-2" When all (or at least one) text elements in "3-2" are updated, these text elements "tournament name", "match name", "country name 1", and "country name" stored in the information management table 13 are updated. 2" is generated.

In addition, as utterance number 3, the utterance content is "Pitch type (curving ball)", the label combination is "5-1-1-12-1", and the trigger label is "5-1-1-12". is defined. Furthermore, as utterance number 4, the following information is displayed: the utterance content is "pitch type (straight)", the label combination is "5-1-1-12-2", and the trigger label is "5-1-1-12". Defined.

This applies to the "pitch type" which is the "group" of the trigger label "5-1-1-12" stored in the information management table 13, and the label "5-1-1-12-1" that belongs to this. The label stored in the information management table 13 when the text element "Curving ball" of the "Item" or the text element "Straight" of the "Item" of the label "5-1-1-12-2" is updated. Indicates that an explanatory audio text consisting of the text element "Curving Ball" with the label "5-1-1-12-1" or the text element "Straight" with the label "5-1-1-12-2" is to be generated. There is.

In addition, the trigger label is not the label "5-1-1-12-1" or "5-1-1-12-2" in the 1st to 5th columns, but the label "5-1-1" in the 1st to 4th columns. -12'' consists of the text element ``curving ball'' with the label ``5-1-1-12-1'' and the text element ``5-1-1-12-2'' from the information management table 13. This is because only one of the updated text elements "straight" is read out, and both text elements are not read out at the same time.

Also, as utterance number 5, the utterance contents are "defensive position", "word (direction)", and "batting result (hit)", and the label combinations are "4-1-3-20-1" and "5-1-". 4-41-1" and "5-1-1-21-1", and each information with a trigger label of "5-1-1-21-1" is defined.

This happens when the text element "Hit" of the trigger label "5-1-1-21-1" stored in the information management table 13 is updated ("Hit" is stored). ), the text elements of the labels "4-1-3-20-1", "5-1-4-41-1" and "5-1-1-21-1" stored in the information management table 13 This shows that an explanatory audio text consisting of "defense position", "word (direction)", and "batting result (hit)" is generated.

In this example of utterance number 5, in addition to the text elements "defense position" and "batting result (hit)" obtained from information source 2, there is a fixed character string between them that is a text element that is not obtained from information source 2. (In this example, "word (direction)") is added. As a result, for example, the explanatory audio text "hit in the left direction" can be generated, and the explanatory audio text with content that is easy for the viewer 100 to understand can be generated, and the viewer 100 may feel uncomfortable when viewing the explanatory audio. can be eliminated.

FIG. 6 is a diagram illustrating an example of the utterance definition data defined in the template 14, the labels and text elements stored in the information management table 13, the Json file, and the commentary audio text.

As shown in FIG. 6(1), the utterance definition data of utterance number 1 of the template 14 includes the utterance contents "name of pitcher information" and "pitcher's action", and the label combination "4-1- Each information of "3-9-1" and "5-1-1-11-1" and a trigger label "5-1-1-11-1" are defined. This is similar to the utterance definition data of utterance number 1 shown in FIG.

Here, in the information management table 13, "kamaeta" is newly stored as "pitcher's action (kamaeta)", which is the text element of the label "5-1-1-11-1", and this data is Suppose it has been updated.

At this time, as shown in FIG. 6(2), at the time of update, the information management table 13 contains the "pitcher information name" of the text element with the label "4-1-3-9-1". , "Pitcher Suzuki" is stored. In addition, in the information management table 13, it is assumed that "kamaeta" is stored as "pitcher's action (kamaeta)" of the text element with the label "5-1-1-11-1". do.

In this case, in accordance with the utterance definition data of utterance number 1 defined in the template 14, the update of the text element to which the trigger label "5-1-1-11-1" is attached is determined in the information management table 13, and the label The text element "Pitcher Suzuki" to which "4-1-3-9-1" is attached and the text element "Kamaeta" to which the label "5-1-1-11-1" is attached are read out.

Then, the Json file and explanatory audio text shown in FIGS. 6(3) and (4), which will be described later, are generated, and the mobile terminal 5 plays "Pitcher Suzuki Kameta" as the explanatory audio audio file.

As mentioned above, the utterance definition data of utterance number 1 of the template 14 shown in FIG. -1-3-9-1" and "5-1-1-11-1", the trigger label is "5-1-1-11-1", and this utterance definition data allows the text element to be An explanatory audio text consisting of "name of pitcher information" and "actions of pitcher" is generated.

On the other hand, in the utterance definition data of utterance number 1, for example, once out of 5 updates, explanatory audio text consisting of the text elements ``name of pitcher information'' and ``pitcher's motion (kamaeta)'' is generated. , 4th inning, additional information may be defined so that an explanatory audio text consisting only of the text element "Pitcher's action" is generated.

<Update monitoring unit 15>
FIG. 7 is a flowchart showing an example of processing by the update monitoring unit 15. The update monitoring unit 15 receives the trigger label from the reading unit 16 and monitors whether the text element to which the trigger label has been assigned in the information management table 13 has been updated (step S701). Note that the trigger label is read from the template 14 by the reading unit 16 and output from the reading unit 16 to the update monitoring unit 15.

If the update monitoring unit 15 determines in step S701 that the text element of the trigger label has not been updated (step S701: no update), it continues the process of step S701.

On the other hand, if the update monitoring unit 15 determines in step S701 that the text element of the trigger label has been updated (step S701: updated), it outputs the presence of update and the trigger label to the reading unit 16 (step S702). The process shown in FIG. 7 is performed for each utterance defined in the template 14 (for each utterance definition data).

FIG. 8 is a diagram illustrating a processing example of the update monitoring unit 15. FIG. 8(1) shows a processing example for the utterance with utterance number 1 of the template 14 shown in the examples of FIGS. 5 and 6. The update monitoring unit 15 inputs the trigger label “5-1-1-11-1” from the reading unit 16. Then, in the information management table 13, the text element "Pitcher's action (kamaeta)" with the trigger label "5-1-1-11-1" is updated, and "kamaeta" is stored. (See α in Figure 8).

Then, the update monitoring unit 15 determines whether the timestamp of the text element to which the trigger label "5-1-1-11-1" has been added in the information management table 13 has been updated. )” is updated, and outputs the update and the trigger label “5-1-1-11-1” to the reading unit 16. In this case, the update monitoring unit 15 updates the text element itself of the trigger label "5-1-1-11-1" (from a state where nothing is stored in the area of "pitcher's motion") to " It may also be possible to determine whether the "pitcher's motion (kameta)" has been updated by determining whether the "pitcher's action (kamaeta)" has been updated due to a change to the stored state.

As a result, text elements "Pitcher Suzuki" and "Kamaeta" are obtained corresponding to the utterance contents "Pitcher information name" and "Pitcher's action (Kamaeta)" defined in utterance number 1 of template 14. .

FIG. 8B shows an example of processing for the utterance with utterance number 4 of the template 14 shown in the example of FIG. The update monitoring unit 15 inputs the trigger label “5-1-1-12” from the reading unit 16. Then, in the information management table 13, the text to which the labels "5-1-1-12-1" and "5-1-1-12-2" corresponding to the trigger label "5-1-1-12" are attached The elements "Pitch Type (Changing Ball)" and "Pitch Type (Straight)" are updated, but nothing is stored in the "Pitch Type (Changing Ball)" area, and the "Pitch Type (Straight)" area is updated. Assume that "straight" is stored in (see β in FIG. 8). Or, label "5-1-1-12" among labels "5-1-1-12-1" and "5-1-1-12-2" corresponding to trigger label "5-1-1-12" Assume that the timestamp of the text element "Pitch Type (Straight)" with "-2" is updated, and "Straight" is newly stored in the "Pitch Type (Straight)" area.

Then, the update monitoring unit 15 displays the labels "5-1-1-12-1" and "5-1-1-12-2" that correspond to the trigger label "5-1-1-12" in the information management table 13. The text elements "Pitch type (Curving ball)" and "Pitch type (Straight)" assigned to the label "5-1-1-12-2" are updated. )", the presence of an update is determined, and the presence of an update and the trigger label "5-1-1-12" are output to the reading unit 16.

As a result, the text element "straight" is obtained in response to the utterance content "pitch type (straight)" defined in utterance number 4 of the template 14. In this case, by determining the update based on the time stamp, it is possible to continuously determine whether the text element "STRAIGHT" has been updated even if the text element "STRAIGHT" has been updated continuously.

<Reading unit 16>
FIG. 9 is a flowchart showing a processing example of the reading unit 16. The reading unit 16 reads utterance definition data (utterance number, utterance content, label combination, and trigger label) for each utterance from the template 14 (step S901). Then, the reading unit 16 outputs the trigger label for each utterance to the update monitoring unit 15 (step S902).

The reading unit 16 determines whether or not the presence of update has been input from the update monitoring unit 15 (step S903). If the reading unit 16 determines in step S903 that the presence of update has not been input (step S903: N), it continues the process of step S903.

On the other hand, if the reading unit 16 determines in step S903 that the presence of update has been input from the update monitoring unit 15 (step S903: Y), the reading unit 16 identifies the utterance corresponding to the trigger label input together with the presence of update. Then, the reading unit 16 reads out the label and the text element to which the label is attached from the information management table 13 for each combination of labels (each label) in the utterance (step S904).

Here, the reading unit 16 stores a plurality of text elements of the same type of label (same type label (label with the same numerical value in the fourth and fifth columns)) as the label (read target label) in the information management table 13. If so, one label and text element stored first among these are read out (step S905). Similar labels are labels that have the same numerical values in the 4th and 5th columns (groups and items) of the label to be read, as well as the same numerical values in the 4th and 5th columns (groups and items) of the label to be read; A label whose first column has a numerical value different from the type of label. A specific example will be explained later with reference to FIG.

The reading unit 16 also detects that in the information management table 13, text elements of labels defined in the template 14 (labels to be read) are not stored, and only text elements of labels of the same type other than the labels to be read are stored. If so, read labels and text elements of the same type other than the label to be read.

As a result, even if the explanatory audio production device 1 is unable to acquire data (data related to the labels defined in the template 14) from the main information source 2, data (data related to the labels defined in the template 14) can be obtained from other information sources 2. When data related to a label (same type label) of a different type of information source 2 in the first column in the label defined in the template 14 is acquired, the same type label and text element are read from the information management table 13, and this is reflected. It is possible to generate explanatory audio text. That is, the mobile terminal 5 can reproduce an audio file of explanatory audio in which data acquired from an information source 2 other than the main information source 2 is reflected.

FIG. 10 is a diagram illustrating an example of reading out text elements (step S905). The label "5-1-1-11-1" is defined in the label combination of the template 14, and in the information management table 13, the 4th and 5th columns of the label are 2 as text elements of "11-1". Assume that two text elements are stored.

One is the text element "Kamaeta" TA with the label "5-1-1-11-1", its timestamp is t1, and the other is the text element with the label "4-1-1-11-1" Assume that the TB is "Kamaeda" and its timestamp is t2. It is assumed that t1<t2, and t2 is closer to the current time. The timestamp indicates the time when the label and text element were stored in the information management table 13.

These labels have the same "11-1" in the 4th and 5th columns, the label "5-1-1-11-1" is the label to be read, and the label "5-1-1-11-1" and “4-1-1-11-1” are similar labels. Both labels are the same in that the utterance content corresponding to "11-1" in the 4th and 5th columns of the label is "kameta", and the utterance content corresponding to "11-1" in the 1st to 3rd columns of the label is "5-1-1" and "4". The difference is that the type of information source 2 shown in the first column in ``-1-1'' is ``input tool'' and ``image analysis tool.''

In this case, the reading unit 16 reads the same label “5-1-1-11-1” and Regarding "4-1-1-11-1", compare the timestamps t1 and t2 of labels "5-1-1-11-1" and "4-1-1-11-1", and The oldest (past) timestamp t1 is specified. Then, the reading unit 16 reads out the label "5-1-1-11-1" of the timestamp t1 stored first and its text element "Kamaeta" TA.

Thereby, the mobile terminal 5 can reproduce the audio file of the explanatory audio corresponding to the text element stored first, so that real-time performance matching the video can be realized.

Returning to FIG. 9, the reading unit 16 reads out the text elements according to the preset rules for the labels and text elements read from the information management table 13 according to the "competition event" indicated by the numerical value in the second column of the label. (step S906).

For example, as the utterance definition data of the template 14, the utterance content "score 1, word (pair), score 2" and the corresponding combination of labels, etc. are defined, and the reading unit 16 reads the label from the information management table 13. "15" is read out as the text element "score 1" of "4-2-1-5-1" (the number "2" in the second column indicates that the "competition" is "tennis"). , it is assumed that "15" is read out as the text element "score 2" of the label "4-2-1-5-2".

Regarding the utterance, the reading unit 16 determines that the "competition" is "tennis" according to a preset rule, and reads the text element "15", the text element "pair", and the text element that constitute the utterance content. "15" (in this case, the explanatory audio text is "15 to 15") is modified to the text element "fifteen" and the text element "all" (in this case, the explanatory audio text is "fifteen all").

The reading unit 16 also retrieves the text element of the label “4-4-1-5-1” (the number “4” in the second column indicates the “competition event” of “badminton”) from the information management table 13. Assume that "15" is read out as "score 1" and "15" is read out as text element "score 2" of label "4-4-1-5-2".

In this case, the reading unit 16 determines that the "competitive event" is "badminton" in accordance with the preset rules for the utterance, and the text element "15" and the text element "pair" that constitute the utterance content. and text element "15" (in this case, the explanatory audio text is "15 to 15"), text element "15", text element "pair", text element "15" and text element "tie" (in this case, the text for explanatory audio is "15 to 15"). The audio commentary text has been revised to read "15 to 15 tie."

As a result, if the score is the same in multiple sports events, but the phrases used are different, the phrases can be distinguished for each sports event, and an audio file of explanatory audio tailored to the sports event can be played. be able to.

Further, for example, the reading unit 16 retrieves the labels "5-2-1-16-1" to "5-2-1-16-4" and these text elements "Smash" from the information management table 13. ..., and the labels "5-2-1-17-1" and "5-2-1-17-2" and their text elements "Result (success)" and "Result (failure)" are read out and the preset values are read. According to the established rules, it is determined that the ``sports event'' is ``tennis,'' and points are automatically added based on these combinations. Then, for example, when the score is 30 to 15, the reading unit 16 adds a new label and text element "Suzuki" "VS" "Tanaka" "Thirty" "Fifteen" (in this case, the explanatory audio text is "Suzuki vs. Tanaka "Thirty Fifteen").

In this way, for example, if the text element of a player serving (for example, "Suzuki") is specified, the text elements "Suzuki", "Double Fault", "Suzuki", "Service Ace", "Suzuki", "Return Ace", etc. According to the preset rules in the reading unit 16, text elements can be modified or new text elements can be generated.

Returning to FIG. 9, the reading unit 16 outputs one or more labels and one or more text elements corresponding thereto to the format converting unit 17 for each utterance (step S907).

<Format converter 17>
FIG. 11 is a flowchart showing an example of processing by the format conversion unit 17. The format conversion unit 17 inputs one or more labels and one or more text elements corresponding thereto for each utterance from the reading unit 16 (step S1101).

The format conversion unit 17 assigns an ID for identifying a Json file, which will be described later, to each utterance (step S1102). Then, the format conversion unit 17 sets the playback time by taking into account the delay due to the time of the speech synthesis process by the speech synthesis device 3, etc., based on the time when the Json file is generated in step S1104, which will be described later. S1103). The reproduction time is the time when the mobile terminal 5 reproduces the audio file of the commentary audio, and may be reset by the subsequent order discard control unit 19 and the distribution device 4 shown in FIG. 1.

The format conversion unit 17 generates, for each utterance, a Json file that includes an ID, playback time, one or more labels, and one or more text elements corresponding thereto, according to a preset data format. Then, the format converter 17 outputs the Json file for each utterance to the text generator 18 and the order discard controller 19 (step S1104).

6(1) to (3), the reading unit 16 retrieves from the information management table 13 the label "4-1-3-9-1" and the corresponding text element "Pitcher Suzuki", and The label "5-1-1-11-1" and the corresponding text element "kamaeta" are read out, and the format conversion unit 17 converts the utterance into a label "4-1-3-9-1". and the corresponding text element "Pitcher Suzuki", as well as the label "5-1-1-11-1" and the corresponding text element "Kamaeta" are input. Then, the format conversion unit 17 assigns an ID to the utterance and sets a playback time.

As shown in FIG. 6(3), the format conversion unit 17 inputs the ID "000...2724", the playback time "2021-03-23...2233705Z", and the first label "4-1-3-". 9-1'' and the text element ``Pitcher Suzuki'', as well as the second label ``5-1-1-11-1'' and the text element ``Kamaeta'' are generated.

Thereby, data in different data formats acquired from a plurality of information sources 2 can be unified, and a Json file that does not reflect the identity of the information sources 2 is generated.

<Text generation unit 18>
FIG. 12 is a flowchart showing an example of processing by the text generation unit 18. The text generation unit 18 receives a Json file for each utterance from the format conversion unit 17 (step S1201).

The text generation unit 18 sequentially extracts one or more text elements from the Json file (step S1202), and generates explanatory audio text consisting of one or more text elements (step S1203).

In the example in Figure 6, the text elements "Pitcher Suzuki" and "Kamaeta" are extracted from the Json file shown in Figure 6 (3), and the explanatory audio text "Pitcher Suzuki Kameta" is extracted as shown in Figure 6 (4). " is generated.

Returning to FIG. 12, the text generation unit 18 gives the Json file (to the utterance) a unique utterance ID for identifying the Json file (utterance) (step S1204). Then, the text generation unit 18 calculates the number of characters in the explanatory audio text generated in step S1203, and performs a predetermined calculation process based on the number of characters to create an audio file (wav (Waveform Audio File Format) file) of the explanatory audio text. The time length is calculated (step S1205). Since the process for calculating the time length of an audio file from the number of characters is well known, the description thereof will be omitted here.

The text generation unit 18 outputs the utterance ID and time length to the order discard control unit 19 (step S1206), and outputs the utterance ID and explanatory audio text to the speech synthesis device 3 (step S1207).

Thereby, the speech synthesis device 3 can manage the audio file of the explanatory audio text using the utterance ID. Further, the subsequent order discard control unit 19 can reset the playback time of each utterance using the time length.

Note that although the text generation unit 18 generates the explanatory voice text based on the Json file input from the format conversion unit 17, it also inputs one or more text elements for each utterance from the reading unit 16, and generates the explanatory audio text based on the Json file input from the format conversion unit 17. Audio text may also be generated.

<Order discard control unit 19>
FIG. 13 is a flowchart showing a processing example of the order discard control unit 19. The order discard control unit 19 determines whether there is an input from the format conversion unit 17 (whether it is input timing or not) (step S1301). If the order discard control unit 19 determines in step S1301 that there is an input (step S1301: Y), the process proceeds to step S1302. On the other hand, if the order discard control unit 19 determines in step S1301 that there is no input (step S1301: N), the process proceeds to step S1308.

Moving from step S1301 (Y), the order discard control unit 19 inputs the Json file from the format conversion unit 17, and inputs the utterance ID and time length corresponding to the Json file from the text generation unit 18 (step S1302).

The order discard control unit 19 extracts the playback time and one or more labels from the input Json file, generates utterance data consisting of the utterance ID, playback time, one or more labels, and time length, and creates an array of utterance data. It is added to the last part (step S1303). As a result, the utterance data corresponding to the text element updated in the information management table 13 is added to the array. Note that the utterance data may include an utterance ID and a Json file.

Here, the array is composed of one or more utterance data for each utterance, that is, one or more utterance data corresponding to one or more audio files that will be uttered as explanatory audio. If there is no explanatory audio to be uttered, no utterance data exists in the array. Every time a Json file is input from the format conversion unit 17, speech data corresponding to the Json file is added to the array. Further, the utterance data in the array is discarded by processing such as step S1306, which will be described later.

The order discard control unit 19 assigns priorities of "immediate" > "semi-immediate" > "regular" > "other" based on the "presentation timing" in the third column of labels for the plurality of utterance data in the array. The order of the plurality of utterance data in the array is determined and rearranged so that the sequence becomes (step S1304).

Here, when the utterance data includes a plurality of labels (utterances), and the third column of even one of the plurality of labels is "3" indicating "regular", The "presentation timing" as speech data is treated as "regular". In addition, when multiple labels do not have the numerical value "3" for "regular" and at least one label has the numerical value "1" for "immediate", the order discard control unit 19 changes the "presentation timing" of the speech data to "immediate". ”. In addition, when multiple labels do not have the numerical value "3" for "regular" and at least one label has the numerical value "2" for "semi-immediate", the order discard control unit 19 changes the "presentation timing" as speech data to " treated as "semi-immediate".

With video and images, multiple pieces of information can be conveyed at the same time, but with audio, even if multiple pieces of information are presented at the same time, it is difficult for the viewer 100 to understand and receive the content. In particular, in an explanation audio service in which the explanation audio production device 1 is used, even if there is a plurality of information to be conveyed at the same time, it is necessary to convey the information one by one in order.

Therefore, by rearranging the order of speech data according to the priority in the process of step S1304 and discarding the speech data in processes such as step S1306, which will be described later, an array consisting of one or more speech data is constructed.

The order discard control unit 19 resets the playback times for the plurality of utterance data in the array (step S1305). Specifically, the order discard control unit 19 resets the reproduction time of the utterance data based on the order in the array and the reproduction time and time length of the utterance data in the array.

For example, assume that the order of the speech data added in step S1303 is determined to be second in the array in step S1304. In this case, the order discard control unit 19 adds the time length of the first utterance data after rearrangement to the playback time of the first utterance data before rearranging the utterance data in the array. The time is reset as the reproduction time of the second determined utterance data after the rearrangement.

Note that although the order discard control unit 19 uses the time length calculated by the text generation unit 18, it is also possible to reset the playback time of the speech data using the time length calculated by the speech synthesis device 3. You can also do this. Since the speech synthesis device 3 calculates the time length when generating the audio file, it can calculate the time length with higher accuracy than the text generation unit 18. Therefore, the order discard control unit 19 can reset the playback time with high accuracy by using the time length calculated by the speech synthesis device 3.

If there is a plurality of utterance data with the same type of label in the fourth and fifth columns in the array, the order discard control unit 19 discards the utterance data with the older playback time (step S1306).

Specifically, the order discard control unit 19 identifies speech data whose fourth and fifth columns include labels with the same numerical value (speech data with the same type of label) from a plurality of speech data in the array. Then, the order discard control unit 19 leaves the utterance data including the latest reproduction time among the plurality of identified utterance data, and leaves one or more utterance data including old reproduction time (past than the latest reproduction time). utterance data is discarded. In this case, the order discard control unit 19 resets the playback time for the utterance data in the array after the discard process, using the same process as step S1305.

As a result, in step S1309, which will be described later, text elements with similar labels are updated in the information management table 13 during the waiting state until the utterance ID and playback time of the first utterance data in the array are output. When data is added to the array, the utterance data including the old playback time is discarded, and the utterance data including the latest playback time, that is, the audio file of the explanatory audio corresponding to the added utterance data is played back on the mobile terminal 5. It will be done. Therefore, when playing an audio file of commentary audio that matches the video, it is possible to present the commentary audio that reflects the contents of the latest match situation to the viewer 100, and to realize real-time performance that matches the video. Can be done.

Regarding the utterance data in the array, the order discard control unit 19 discards the utterance data for which a certain period of time (a preset time) has passed since it was added to the array (step S1307), and moves to step S1308. In this case as well, the order discard control unit 19 resets the playback time for the utterance data in the array after the discard process, using the same process as step S1305.

The order discard control unit 19 moves from step S1301 (N) or step S1307 and determines whether there is an output from the order discard control unit 19 (whether it is the output timing or not) (step S1308). .

If the order discard control unit 19 determines in step S1308 that there is an output (step S1308: Y), the process proceeds to step S1309. On the other hand, if the order discard control unit 19 determines in step S1308 that there is no output (step S1308: N), it ends the process and restarts the process from step S1301.

The order discard control unit 19 moves from step S1308 (Y), extracts the utterance ID and playback time from the first utterance data in the array, and outputs the utterance ID and playback time to the distribution device 4 (step S1309). . Then, the order discard control unit 19 discards the first utterance data in the array (step S1310).

FIG. 14 is a diagram illustrating the transition of utterance data in an array in the order discard control unit 19. Assume that the array (A) is configured. The beginning of this array is the utterance data <a> of "Pitcher name (Pitcher information name) + Kamaeta (Pitcher's action (Kamaeta))", and the "Presentation timing" in the third column of labels is "Immediate". ([1]). This array also includes the utterance data <p> of "game name + innings + score (score 1, score 2)", and the "presentation timing" in the third column of the label is "regular" ([ 3]). The explanatory audio text for the utterance data <p> is "X vs. Y, bottom of the 6th inning, 7 to 0."

As shown in (B), the order discard control unit 19 outputs the utterance ID and playback time of the first utterance data <a> in the array in step S1309, and discards the utterance data <a> in step S1310. . As a result, the utterance data <a> is discarded from the array (A), and the array (C) is constructed.

Then, as shown in (D), in step S1302, the order discard control unit 19 creates a Json file of the utterance data <b> of "batter name (name of batter information) + hit (batter's action)", " A Json file of speech data <c> of "ball speed" and a Json file of speech data <d> of "defense position + direction (word (direction)) + hit (batting result (hit))" are input. The "presentation timing" in the third column of the label included in speech data <b> is "immediate" ([1]), and the "presentation timing" in the third column of the label included in speech data <c> is "immediate" ([1]). "Semi-immediate" ([2]). Further, the "presentation timing" in the third column of the labels included in the utterance data <d> is "immediate" ([1]).

As shown in (E), in step S1304, the order discard control unit 19 selects "immediate">"semi-immediate">based on the "presentation timing" in the third column of labels for the plurality of utterance data in the array. The order of the utterance data is determined and rearranged so that the priority is "regular">"other", and in step S1305, the playback time of the utterance data is reset to form the array (F).

The beginning of the array in (F) is speech data <b>, and the "presentation timing" in the third column of labels is "immediate" ([1]). The second is speech data <d>, and the "presentation timing" in the third column of labels is "immediate" ([1]). The third item is speech data <c>, and the “presentation timing” in the third column of labels is “semi-immediate” ([2]). For the fourth and subsequent utterance data, the "presentation timing" in the third column of the label is "semi-immediate" ([2]), "regular" ([3]), or "other" ("4"). Suppose there is.

Then, as shown in (G), the order discard control unit 19 outputs the utterance ID and playback time of the utterance data <b> at the beginning of the array in step S1309, and in step S1310, the order discard control unit 19 outputs the utterance ID and reproduction time of the utterance data <b> at the beginning of the array. Discard. As a result, speech data <b> is discarded from the array (F). Further, the order discard control unit 19 outputs the utterance ID and playback time of the utterance data <d> at the beginning of the array in step S1309, and discards the utterance data <d> in step S1310. As a result, the utterance data <d> is discarded from the array.

Then, as shown in (H), in step S1302, the order discard control unit 19 creates a Json file of the utterance data <e> of "throwing the ball from left field to third base" and "game name + inning + score". Input the Json file of speech data <f>. The "presentation timing" in the third column of the label included in the utterance data <e> is "immediate" ([1]), and the "presentation timing" in the third column of the label included in the utterance data <f> is "immediate" ([1]). Periodic” ([3]).

Here, the labels of the utterance data <f> and the utterance data <p> are of the same type. The audio commentary text for the aforementioned utterance data <p> is "X vs. Y, bottom of the 6th inning, 7-0", whereas the audio commentary text for the updated utterance data <f> is "X vs. Y, 6th inning". Back 7 to 1”.

As shown in (I), in step S1304, the order discard control unit 19 selects "immediate">"semi-immediate">based on the "presentation timing" in the third column of labels for the plurality of utterance data in the array. The order of the utterance data is determined and rearranged so that the priority is "regular">"other", and the playback time of the utterance data is reset in step S1305.

In this case, the order discard control unit 19 sets the "presentation timing" to "immediate" for the updated speech data <f> whose "presentation timing" is "regular" and determines the order of the speech data. Sort by. Then, the order discard control unit 19 sets the "presentation timing" of the utterance data <f> to "immediately" for the period from the next to the last utterance data of the utterance data <f> for which the "presentation timing" is set to "immediate". The utterance data <f> is inserted at a preset time interval while leaving the setting as ``periodic''.

As a result, the explanatory audio text "X vs. Y, bottom of the 6th inning, 7 vs. 1" for the utterance data <f> is generated immediately, and thereafter is generated periodically.

Then, as shown in (J), in step S1306, the order discard control unit 19 determines that there is speech data <f><p> with the same type of label in the array, and that the speech data <p> is the same as the speech data <f>. Since it has been updated, the utterance data <p> other than the newly added utterance data <f> are discarded. As a result, the utterance data <p> corresponding to the explanatory audio text "X vs. Y, bottom of the 6th inning, 7-0" with the oldest playback time is discarded from the array, and the explanatory audio text "X vs. Y 6" with the latest playback time is discarded. The utterance data <f> corresponding to ``7 to 1 at the back of the inning'' remains. Furthermore, in step S1307, the order discard control unit 19 discards the utterance data <c> after a certain period of time has passed since the arrangement. As a result, the array (K) is constructed.

The beginning of the array in (K) is the utterance data <e>, and the "presentation timing" in the third column of labels is "immediate" ([1]). The second is speech data <f>, and the "presentation timing" in the third column of labels is "immediate" ([1]). Regarding this utterance data <f>, since the "presentation timing" when inputting the Json file is "regular" ([3]), the utterance data <f'> is displayed at a preset time interval. is inserted into the array as .

As a result, the commentary audio text "X vs. Y, bottom of the 6th inning, 7 to 1" for the utterance data <f> is generated immediately and periodically, and the viewer 100 can listen to this commentary audio. You can listen to it when it updates, and then periodically until it updates.

Here, for utterance data that includes multiple labels, if the utterance data includes a label whose "presentation timing" is "regular" and further includes a label whose "immediate" or "semi-immediate", the "presentation timing" of the utterance data is treated as "periodic".

In this case, as described in (I) above, when configuring a new array, the order discard control unit 19 uses the updated utterance data (newly input Json file) whose "presentation timing" is "regular". utterance data), set the ``presentation timing'' in the first utterance data to ``immediate'' or ``semi-immediate'' (if other utterance data in the array contains the label ``immediate'' and is set to ``semi-immediate''). If the utterance data does not contain a label, set it to "immediate"; if the other utterance data contains a label of "semi-immediate" and does not contain a label of "immediate," set it to "semi-immediate"), and Sort the utterance data in the array. Then, the order discard control unit 19 determines the order and arranges the utterance data in the array so that the "presentation timing" of the subsequent utterance data is set to "regular" and at a preset time interval. Change.

As described above, according to the explanatory audio production device 1 according to the embodiment of the present invention, the analysis unit 11 inputs data from a plurality of information sources 2 and inputs data according to the preset data format of the information sources 2. Extract text elements by analyzing data.

The storage unit 12 assigns a label to a text element and stores the labeled text element together with a time stamp in the information management table 13.

The update monitoring unit 15 monitors whether the text element of the trigger label included in the utterance definition data defined in the template 14 has been updated in the information management table 13.

The reading unit 16 reads one or more text elements of a combination of labels (one or more labels) included in the utterance definition data from the information management table 13 for the utterance of the updated trigger label of the text element.

The format conversion unit 17 converts the format of one or more labels and one or more text elements corresponding thereto into a Json file including playback time.

The text generation unit 18 extracts one or more text elements from the Json file, generates explanatory audio text, assigns an utterance ID to the Json file, and generates the audio based on the number of characters in the explanatory audio text. Calculate the time length of the file.

When a Json file is input from the format conversion unit 17, the order discard control unit 19 extracts the playback time and one or more labels from the Json file, and extracts the utterance ID, playback time, one or more labels, and time length. , and the order of the utterance data is determined based on the presentation timing priority included in the label to form an array. Furthermore, the order discard control unit 19 resets the playback time of the speech data based on the configured order of the array, the playback time and time length of the speech data.

The order discard control unit 19 outputs the utterance ID and playback time of the first utterance data in the array, and discards the utterance data. In addition, when there is a plurality of speech data with the same type of label in the array, the order discard control unit 19 discards the speech data with the oldest playback time, and also discards the speech data after a certain period of time has elapsed.

As a result, the explanatory audio text to be uttered is generated according to the utterance definition data for each utterance defined in the template 14, and the order of utterances is determined based on the priority of "presentation timing" in the third column of the label. The playback time of the utterance is reset.

By playing the audio file of the explanatory audio text generated in this way at the playback time, the explanatory audio can be provided to the viewer 100 in real time in conjunction with the live broadcast of the sports program being distributed live. can.

In addition, since data is acquired from multiple information sources 2 and text elements extracted from the data are stored in the information management table 13 along with a label that unifies the data formats of the multiple information sources 2, the information management table 13 can be centrally managed. In other words, it is possible to realize a highly versatile explanatory audio production and distribution system 10.

In this way, by centrally managing the information management table 13 using a plurality of information sources 2, explanatory audio text for each utterance without overlapping text elements is generated while ensuring real-time performance.

If it is desired to increase the number of utterances of this explanatory voice text, it is only necessary to add the utterance definition data of the utterances to the template 14, so it is possible to realize a highly extensible explanatory voice production and distribution system 10.

Therefore, it is possible to utilize data from a plurality of information sources 2 and to provide explanatory audio with high expandability and versatility in real time.

Although the present invention has been described above with reference to the embodiments, the present invention is not limited to the embodiments described above, and can be modified in various ways without departing from the technical concept thereof.

For example, although the format conversion unit 17 is configured to generate a Json file, it may also generate files other than Json files. In the present invention, the files input by the text generation section 18 and the order discard control section 19 are not limited to Json files, but may be files in accordance with other data formats.

In addition, in the explanatory audio production and distribution system 10 shown in FIG. did.

On the other hand, the distribution device 4 may distribute the explanatory audio text and the reproduction time generated by the explanatory audio production device 1 to the mobile terminal 5 instead of the audio file and the reproduction time. In this case, the commentary audio production and distribution system 10 does not need to include the speech synthesis device 3, and the mobile terminal 5 generates an audio file by receiving the commentary audio text and performs speech synthesis processing, and plays the audio file. Play at the time.

Furthermore, as shown in FIG. 5, the template 14 defines utterance content, label combinations, and the like. On the other hand, information such as a subtitle may be defined in the template 14. In this case, a highly versatile item such as a caption is prepared as the fifth column of labels, and the storage unit 12 stores text elements of information given such a label in the information management table 13. . Then, the text generation unit 18 generates explanatory audio text including text elements such as telops. As a result, even if it is not known what kind of information should be presented as commentary audio, text for commentary audio that includes text elements such as captions can be generated, and the rate of providing explanations for programs can be improved. can.

For example, text elements of data input from the information source 2-5 of the input tool are stored in the information management table 13 together with labels such as captions. As a result, the character information inputted by the operator of information source 2-5 of the input tool will be reflected in the commentary audio text, and the audio file of the commentary audio in which the text information is reflected will be played. Become.

Note that a normal computer can be used as the hardware configuration of the explanatory audio production device 1 according to the embodiment of the present invention. The explanatory audio production device 1 is constituted by a computer equipped with a CPU, a volatile storage medium such as a RAM, a nonvolatile storage medium such as a ROM, an interface, and the like.

Each of the analysis section 11, storage section 12, information management table 13, template 14, update monitoring section 15, reading section 16, format conversion section 17, text generation section 18, and order discard control section 19 provided in the explanatory audio production device 1 The functions are realized by causing the CPU to execute programs in which these functions are written.

These programs are stored in the storage medium, and are read and executed by the CPU. Additionally, these programs can be stored and distributed in storage media such as magnetic disks (floppy (registered trademark) disks, hard disks, etc.), optical disks (CD-ROMs, DVDs, etc.), semiconductor memories, etc., and can be distributed via networks. You can also send and receive messages.

1 Explanation audio production device 2 Information source 3 Speech synthesis device 4 Distribution device 5, 105 Mobile terminal 10 Explanation audio production and distribution system 11 Analysis section 12 Storage section 13 Information management table 14 Template 15 Update monitoring section 16 Reading section 17 Format conversion section 18 Text generation unit 19 Order discard control unit 100 Viewer 101 Broadcast transmission device 102 Broadcast reception device 103 Explanation audio production and distribution device 104 Application server

Claims (11)

In a commentary audio production device that generates text for commentary audio for each utterance of a live sports program,
a template in which utterance definition data including one or more labels corresponding to the one or more text elements when the text is composed of one or more text elements is defined for each of the utterances;
an information management table in which the text elements are stored;
Analysis that extracts the text element from the data by inputting data corresponding to the match situation of the sports program from each of a plurality of information sources and analyzing the data according to a preset data format of the information source. Department and
a storage unit that assigns a label including a priority of the timing of presenting the utterance of the text element to the text element extracted by the analysis unit, and stores the text element to which the label has been assigned in the information management table; and,
an update monitoring unit that monitors whether or not the text element stored in the information management table has been updated, and outputs the label given to the text element when it is determined that the text element has been updated;
Regarding the utterance of the utterance definition data including the label output by the update monitoring unit, one or more corresponding texts to which one or more labels included in the utterance definition data are attached are determined from the information management table. a reading unit that reads the element and outputs one or more labels of the utterance and one or more text elements corresponding thereto;
a format conversion unit that converts the format of one or more labels of the utterance output by the reading unit and one or more text elements corresponding thereto into a file including a predetermined playback time;
a text generation unit that extracts the one or more text elements from the file whose format has been converted by the format conversion unit, generates and outputs the text;
The file for each utterance whose format has been converted by the format converter is input, the order of the utterances is determined based on the priority included in the label of the file for each utterance, and the file for each utterance is converted according to the order. reset the playback time included in the file;
an order discard control unit that outputs the playback time included in the file of the first utterance for which the order has been determined, and discards the file of the first utterance;
An explanatory audio production device characterized by comprising: The explanatory audio production device according to claim 1,
The plurality of information sources include an information source that transmits real-time data according to the match situation of the sports program, and further includes an information source that transmits data of the sports program according to an input operation by an operator, and a source that transmits real-time data according to the match situation of the sports program. The information source includes at least one of an information source that transmits data obtained by analyzing an image of a match situation of a program, and an information source that transmits data obtained by recognizing audio of a match situation of the sports program. An explanatory audio production device characterized by: The explanatory audio production device according to claim 1,
In the template, for each utterance, in addition to the one or more labels, one of the one or more labels is defined as a trigger label,
The update monitoring unit includes:
An explanatory audio production device characterized in that it monitors whether or not the text element to which the trigger label is attached has been updated in the information management table, and outputs the trigger label when it is determined that the text element has been updated. The explanatory audio production device according to claim 1,
The label is composed of respective numerical values indicating the type of information source, the competition type of the sports program, the priority, the group to which the text element belongs, and the item within the group,
When reading the corresponding one or more text elements to which the one or more labels included in the utterance definition data are attached, the label is set as a read target label, and in addition to the read target label, the read target label is configured. A label to which the group to which the text element belongs and the item in the group are the same, and the type of the information source is different is considered a homogeneous label,
The reading section is
When the information management table stores a plurality of text elements to which the same type of label is attached,
An explanatory audio production device characterized in that the text element stored first among a plurality of text elements to which the same kind of label is given is read out from the information management table. The explanatory audio production device according to claim 4,
The reading section is
When the information management table does not store a text element to which the read target label is attached, but stores a text element to which the same kind of label other than the read target label is attached,
An explanatory audio production device characterized in that the text element to which the same kind of label other than the read target label is attached is read out from the information management table. The explanatory audio production device according to claim 1,
In addition to the priority, the label includes the competition type of the sports program,
The reading section is
Regarding the text element read from the information management table, the text element is modified according to the sport event included in the label given to the text element, and one or more labels and one or more texts of the utterance are modified. An explanatory audio production device characterized by outputting an element (if there is a corrected text element, the text element). The explanatory audio production device according to claim 1,
The one or more labels included in the utterance definition data include a label corresponding to a predetermined particle or word,
The reading section is
reading one or more text elements containing the predetermined particle or word from the information management table;
The text generation unit is
An explanatory audio production device characterized by extracting one or more text elements containing the predetermined particle or word from the file, and generating and outputting a text containing the predetermined particle or word. The explanatory audio production device according to claim 1,
The priority included in the label is any one of information indicating immediate, semi-immediate, regular, and other, and the immediate priority is the highest, the semi-immediate is the next highest, and the other is the priority. is the lowest,
The order discard control unit includes:
utterances of a label containing the periodic priority, a first utterance of the label containing the immediate or semi-immediate priority, and a second utterance of the label containing the periodic priority arranged at predetermined time intervals; As an utterance,
Regarding the first utterance, determining the order of the utterances such that the higher the priority, the closer to the beginning the utterances are, and
An explanatory audio production device characterized in that the order of the utterances is determined so that the second utterance is placed after the first utterance at the predetermined time interval. The explanatory audio production device according to claim 1,
The label is composed of a type of information source, a competition type of the sports program, the priority, a group to which the text element belongs, and a numerical value indicating an item within the group,
When reading the corresponding one or more text elements to which the one or more labels included in the utterance definition data are attached, the label is set as a read target label, and in addition to the read target label, the read target label is configured. A label to which the group to which the text element belongs and the item in the group are the same, and the type of the information source is different is considered a homogeneous label,
The order discard control unit includes:
In accordance with the update determination by the update monitoring unit, when a new file including the same kind of label is input and the order of utterances for the files for each utterance is determined, it is determined that there are multiple files including the same kind of label. case,
An explanatory audio production device characterized in that, among a plurality of files including the same kind of label, files other than the new file are discarded. The explanatory audio production device according to claim 1,
The order discard control unit includes:
An explanatory audio production device characterized in that, among the files for each utterance, files for which a preset time has elapsed are discarded. A computer that constitutes a commentary audio production device that generates text for commentary audio for each utterance of sports programs that are being streamed live.
a template in which utterance definition data including one or more labels corresponding to the one or more text elements when the text is composed of one or more text elements is defined for each of the utterances;
an information management table in which the text elements are stored;
Analysis that extracts the text element from the data by inputting data corresponding to the match situation of the sports program from each of a plurality of information sources and analyzing the data according to a preset data format of the information source. Department,
a storage unit that assigns a label including a priority of the timing of presenting the utterance of the text element to the text element extracted by the analysis unit, and stores the text element to which the label has been assigned in the information management table; ,
an update monitoring unit that monitors whether or not the text element stored in the information management table has been updated, and outputs the label given to the text element when it is determined that the text element has been updated;
Regarding the utterance of the utterance definition data including the label output by the update monitoring unit, one or more corresponding texts to which one or more labels included in the utterance definition data are attached are determined from the information management table. a reading unit that reads the element and outputs one or more labels of the utterance and one or more text elements corresponding thereto;
a format conversion unit that converts the format of one or more labels of the utterance output by the reading unit and one or more text elements corresponding thereto into a file including a predetermined playback time;
a text generation unit that extracts the one or more text elements from the file whose format has been converted by the format conversion unit, generates and outputs the text, and
The file for each utterance whose format has been converted by the format converter is input, the order of the utterances is determined based on the priority included in the label of the file for each utterance, and the file for each utterance is converted according to the order. reset the playback time included in the file;
A program for functioning as an order discard control unit that outputs the playback time included in a file of the first utterance whose order has been determined, and discards the file of the first utterance.

Images