JP2008097232A - Voice information retrieval program, recording medium thereof, voice information retrieval system, and method for retrieving voice information - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a voice information retrieval program capable of obtaining a more accurate retrieval result and allowed to be easily used for various databases. <P>SOLUTION: The voice information retrieval program allows a computer to perform: at least a step for converting voice elements recorded in voice information into character strings; a step for counting the reproduction time of the voice information; a step for dividing the character strings in each prescribed time and registering the divided character strings in a database when the reproduction time of the voice information is longer than the prescribed time, and when the reproduction time of the voice information is the prescribed time and less, registering the character strings of the voice information in the database without division; a step for determining a retrieval item for retrieving the database and a retrieving condition of the retrieval item; a step for performing the retrieval of the database according to the retrieving condition; and a step for outputting the result of retrieval. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a voice information retrieval program. More specifically, the present invention relates to a voice information search program and its recording medium, a voice information search system, and a voice information search method.

  With the recent promotion of computerization, various types of electronic information such as text information, image information, audio information, and moving image information are converted into data and used. Then, a user accesses a database in which these pieces of information are stored, and a desired information file can be obtained by searching in the database. In this regard, with the diversification of types of information files registered in a database to be searched, a search method has been developed.

  For example, by inputting a keyword to a voice information file such as a lecture or a news program, a voice information file including the keyword is searched. In that case, it is important how efficiently and accurately the time zone (reproduction time position) in which the desired keyword is issued in the audio information file can be searched. For example, it is the user's time burden to be able to efficiently search for the scene that the user wants to see (the scene where the desired speech is emitted and its playback time position) in movie files and audio information files of movies and lectures over several hours. As well as alleviating, the stress when using search is also reduced.

  As such a speech information retrieval method, a search keyword input by a user is converted into a speech element, and matching with speech stored in a speech information file of a database is performed. Further, Patent Document 1 discloses a technique for searching in consideration of a possibility that a recognition error has occurred when generating a speech element sequence from an input arbitrary word as a speech information search method. Has been.

JP-A-2005-257594.

  However, when searching for voice information, if the voice information file stored in the database is used as a voice element, voice search may not be performed correctly due to the intonation of the voice, the presence or absence of strength, or the presence of rounding. descend. Also, if the amount of information recorded in the audio information file (recording time, recording time, etc.) is large, the desired scene (playback time position) may not be accurately searched from this audio information file. Accuracy is reduced.

  SUMMARY OF THE INVENTION Accordingly, it is a primary object of the present invention to provide an audio information search program that can obtain more accurate search results and can be easily used in various databases.

  First, as a first aspect of the present invention, a step of converting an audio element recorded in audio information into a character string, a step of measuring the reproduction time of the audio information, and a reproduction time of the audio information are longer than a predetermined time If it is long, the character string of the voice information is divided into predetermined time intervals and registered in the database. If the reproduction time of the voice information is less than the predetermined time, the character string of the voice information is not divided. Registering in the database; determining a search item for searching the database; a search condition for the search item; and executing a search for the database according to the search condition; And a voice information retrieval program for causing a computer to execute at least the step of outputting the search result. By causing the computer to execute each of these steps, it is possible to perform voice information search with high search accuracy and high search speed.

  As a second aspect of the present invention, the number of characters constituting the character string is measured, and when the number of characters is longer than the predetermined number of characters, the segmentation is performed. You can also

  Next, in the present invention, in order to perform a search on the database, at least (1) a step of combining two pieces of continuous voice information among the divided pieces of voice information, and (2) the combination. A step of determining whether or not the character specified by the search condition is included in the character strings of the two voice information; (3) each of the voice information in the character strings of the two separated voice information; A voice information retrieval program is provided that causes a computer to execute a step of determining whether a character string includes a character designated by a retrieval condition. By causing the computer to execute each of these steps, it is possible to perform speech information retrieval with high retrieval accuracy even for character strings that are truncated or truncated by separating speech information.

  Subsequently, in the present invention, in order to output the search result, at least (a) a step of designating a recording time position of voice information in which a voice element corresponding to the character designated by the search condition is recorded; b) designating a computer as a reproduction start position a recording time position that goes back a predetermined time before the designated recording time position; and (c) performing audio reproduction from the designated reproduction start position. Provided is a speech information retrieval program characterized by being executed. By making the computer execute these steps, the audio information can be reproduced from a more accurate reproduction position that matches the search condition.

  Furthermore, the present invention provides a computer-readable recording medium on which the program is recorded. The “recording medium” used in the present invention refers to any computer-readable recording medium on which a program is recorded, which is used for program installation, execution, program distribution / distribution, and the like.

  In the present invention, the voice element recorded in the voice information is converted into a character string, the voice information playback time measuring means, and when the voice information playback time is longer than a predetermined time, Means for registering the character string of the voice information into the database divided every predetermined time and registering the database in the database without dividing the character string of the voice information when the reproduction time of the voice information is equal to or less than the predetermined time; A search item for searching the database; a search condition determining means for the search item; a means for executing a search for the database according to the search condition; and a search result output means; A speech information retrieval system comprising at least By using such a voice information search system, it is possible to construct a voice information search system that has high search accuracy and high search speed.

  And, in the present invention, when the sound element recorded in the sound information is converted into a character string, the procedure for measuring the reproduction time of the sound information, and the reproduction time of the sound information is longer than a predetermined time, A procedure for registering the character string of the voice information in a database divided into predetermined time intervals and registering it in the database without dividing the character string of the voice information when a reproduction time of the voice information is equal to or less than a predetermined time. A search item for searching the database, a search condition for the search item, a procedure for executing a search for the database according to the search condition, and a result of the search And a method for outputting speech information. According to this speech information retrieval method, it is possible to perform speech information retrieval that has high retrieval accuracy and high retrieval speed.

  It should be noted that “audio information” in the present invention means any information in which at least an audio element is recorded, and includes, for example, moving image information, unless otherwise specified. Similarly, the “audio information file” in the present invention includes at least a moving image information file in which an audio element is recorded, unless otherwise specified. In the present invention, steps, procedures, means, and the like are only required to achieve the object of the present invention, and the order of performing these steps is not limited within the scope of the present invention.

  According to the voice information search program according to the present invention, voice information can be searched with high search accuracy, and voice information search can be performed at a high search speed.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings. In addition, embodiment described below illustrates suitable embodiment of this invention and this invention is not interpreted narrowly by this.

  FIG. 1 is a conceptual diagram for explaining a voice information retrieval program according to the present invention.

  As shown in FIG. 1, a speech information retrieval program according to an embodiment of the present invention includes a speech information file text conversion function 1, a speech information file playback time measurement function 2, a database registration function 3, and search condition determination. The function 4, the search processing function 5, and the search result output function 6 are executed by at least a computer, and the audio information file is stored in the database 7. Each function will be described below.

  The voice information file text conversion function 1 is a function for converting a voice element input to the voice information file into a character string. The audio information file reproduction time measurement function 2 is a function for measuring the reproduction time of the audio information file. The database registration function 3 is a function for registering an audio information file in the database 7. The search condition determination function 4 is a function for determining a search condition for extracting a desired audio information file. The search processing function 5 is a function for executing a search on the database 7 in accordance with the search condition determined by the search condition determination function 4. The search result output function 6 is a function for outputting the search result obtained by the search processing function 5.

  FIG. 2 is a flowchart for registering a database in the first embodiment of the speech information retrieval program according to the present invention. That is, FIG. 2 is a flowchart for registering the audio information file in the database based on the reproduction time.

  First, file information of an audio information file (including a moving image information file; the same applies hereinafter) is input (S1). As the file information, the title name, category, creator, voice language, creation date, etc. of the voice information file are input. As described above, in the present invention, when content information is registered as audio information, filter condition items such as a content name and a category can be input. However, the file information input in this step is not particularly limited, Information necessary for database construction can be input. For example, basic content information may be input. The input method is not particularly limited in the present invention, and can be performed using, for example, a keyboard. Various information input in this way is registered in the database as file information of the audio information file (S2).

  Next, it is determined whether the voice element of the voice information file is Japanese or English (S3). As a result, if it is determined that the audio information file is in Japanese, noise cut processing for Japanese is performed (S4a). If it is determined that the audio information file is in English, noise cut processing for English is performed (S4b). It is desirable to process this noise cut while reproducing the audio information file.

  As described above, in the present invention, the voice information file is reproduced, converted into text using the voice conversion system, and stored in the database. When the reproduced voice is Japanese (see S3, S4a, etc.), the text conversion is displayed in hiragana. Alternatively, display of Japanese conversion is not particularly limited, and katakana display may be used.

  When the reproduced voice is in a foreign language such as English, the text conversion is displayed in the foreign language such as an English character (see S2, S3b, etc.). In the present invention, the target language is not particularly limited, and may be Japanese, English, Chinese, Korean, French, etc., but preferably has a homonym. In this case, in the case of speech information, since words (characters) corresponding to speech can be uniquely converted and determined, higher search accuracy can be obtained.

  In the present invention, steps S1 to S4 are not necessarily performed, and can be appropriately performed from the viewpoint of search items used as search conditions, easy management of audio information files on a database, and the like.

  Next, the audio information file is reproduced (S5), and the audio elements of the reproduced audio information file are converted to text (S6). In the present invention, there is no particular limitation on the method for text-converting a voice element of a voice information file. For example, text conversion can be performed using a voice conversion system such as “Microsoft (registered trademark) R.Net Speech”.

  Then, it is determined whether the reproduction time of the reproduced audio information file is longer or shorter than n seconds (S7). As a result, when it is determined that it is shorter than n seconds, the text information of the voice information file is registered in the database (S8a). The audio information file processed in step S8a is a short audio information file whose reproduction time is a predetermined time n seconds or less.

  As a result of the determination in step S7, if it is determined that it is longer than n seconds, the playback portion of the first n seconds of the audio information file is divided and the text information of the divided audio information file is registered in the database. (S8b). Then, for the remaining portion of the cut audio information file, it is determined again in step S7 whether or not the reproduction time is longer than n seconds, and the audio information file is finally divided in units of n seconds. The audio information file processed in step S8b is an audio information file whose reproduction time is longer than the predetermined time n seconds.

  In other words, in the present invention, the reproduced audio information is converted into text and stored in the database. At that time, by pressing a time stamp at predetermined time intervals (n seconds), it is divided into units. To be recorded. Further, when the audio information has a short information amount equal to or shorter than the predetermined time interval (n seconds), the time stamp is not divided and is recorded as it is (see FIG. 3 and the like).

  In the present invention, the time setting of the predetermined time n seconds in step S7 is not particularly limited, and can be set as appropriate in consideration of the search accuracy, the processing capability of the computer used, the use environment, and the like. For example, when it is desired to increase the search accuracy, the time setting for n seconds may be shortened, and when the search speed is important, the time setting for n seconds may be increased.

  As described above, all audio information files that have finally become the predetermined time n seconds or less are registered in the database.

  FIG. 3 is a conceptual diagram illustrating a basic information table of audio information files registered in the database in the first embodiment (see FIG. 2, S1, etc.). FIG. It is a conceptual diagram explaining the content data of the audio | voice information file registered into the database (refer FIG. 2; S7, S8a, S8b etc.).

  FIG. 3 is an example of file information of an audio information file, and four programs with content IDs “1” to “4” are registered. Information on the titles, categories, and speech languages of the four programs is shown. For example, the content ID “1” is assigned to the first registered program (“homepage creation”) as a number for identifying and identifying the program. The category is “Computer”. Thus, the search accuracy can be further improved by registering the genre or the like of the content program. Then, “Japanese” as the voice language is registered as file information.

  FIG. 4 shows content data of an audio information file registered in the database, and the audio information file is converted into text and divided so as to have a reproduction time of a predetermined time n seconds or less (FIG. 2; (See S8a, 8b, etc.). 4 illustrates the program “homepage creation” to which the content ID “1” in FIG. 3 is assigned and the program “Today's weather in Tokyo” to which the content ID “4” is assigned. The content ID is a number assigned to each program as in FIG. The record ID indicates what number of audio information is reproduced in the program. For example, the record ID “1” is an audio information file reproduced first in the program. The record ID “2” indicates that it is the second audio information file reproduced next to the record ID “1”.

  The time stamp in FIG. 4 indicates the playback time from the beginning of the program. For example, the time stamp “00: 00: 00: 00” indicates that it is 0 seconds from the beginning of the program, and the time stamp “00: 00: 10: 00” is exactly 10 seconds from the beginning of the program. Shows that it was time. That is, in this case, the predetermined time n of each record is 10 seconds (see FIG. 2; S8b etc.).

  On the other hand, in FIG. 4, for the program “Today's weather in Tokyo” to which the content ID “4” is given, the time stamp is “00:00:00” and the record ID is only “1”. The reproduction time of this program is a short program with a total of 8 seconds (see FIG. 2, S8a). Therefore, this program is an audio information file shorter than the predetermined time n (n = 10 seconds), and the record “ID” is a content composed only of “1”.

  Also, the item “text” in FIG. 4 indicates data obtained by converting the voice element of the voice information file into characters. For example, in the record ID “1” of the content ID “1”, the voice of “I'm going to create a homepage from now on, but first I need everything ...” is text-converted. The last “Short ...” is divided into a predetermined time n (n = 10 seconds). The next record ID “2” is a continuation of the record ID “1”, and the text of the record ID “2” must be “... The voice of "Moryo ..." is converted to text.

  FIG. 5 is a flowchart for performing a search process on the database in the first embodiment. That is, FIG. 5 is a flowchart for performing a search process on the database registered in FIG. 2, and shows a search process for searching for an audio information file including a predetermined keyword.

  First, search conditions are input (S9). First, as a search condition item to be input, a category, a language, a text keyword, and the like of the voice information file are listed. In the present invention, the item of the search condition item (see S2 in FIG. 2) input when creating the database can be input. In the present invention, the search condition items are not particularly limited. For example, a category, a language, an included keyword (character string), and the like can be set.

  Next, based on the registered content basic information table, data having content basic information that matches the set conditions such as category and language is searched and extracted (S10). This step is a step of searching based on the basic information (see FIG. 2, FIG. 3, etc.) given to the audio information file.

  As a result of step S10, it is determined whether the number of extraction cases is 0 or 1 or more (S11). As a result, if there is one or more extractions that match the search condition, a content ID list of content IDs of the corresponding data (see FIG. 2 etc.) is created. On the other hand, when the number of extracted cases is 0, “not applicable” is displayed (12b).

  Subsequently, the number of contents ID lists is looped (S13 to S26). That is, all the contents extracted in step S10 are searched. For example, when searching in the database using the category “computer” as a search condition, all the contents belonging to the category “computer” are extracted, but all the records of the content group are subjected to text search.

  The text search is performed in the following steps. First, the count of content data is reset (S14), and a search is made for records with matching content IDs in the same content data table (S15 to 25). First, a search is made with respect to a record group registered by being divided within the content (within the program). For the record ID, the nth ID and the (n + 1) th ID are extracted (S16). For example, records with record IDs “1” and “2” are extracted. Alternatively, records with record IDs “2” and “3” are extracted.

  Next, the texts of the extracted record IDs “n” and “n + 1” are combined (S17). Then, a keyword search is performed on the character strings in the combined record IDs “n” and “n + 1”. In this way, by performing a text search by combining the two previous and subsequent record IDs, it is possible to perform a search even for the words (head and tail) that are separated before and after the record ID. For example, in the example of FIG. 3, by combining the content IDs “2” and “3”, a search is also made for the word “Muryo / U (free)” delimited by the content IDs “2” and “3”. be able to. That is, from the beginning of the content ID “2”, the last character string of the content ID “3” “If you do n’t get it, it ’s free and you can create it” (see FIG. 3). Search for. As a result, when the corresponding keyword is included, the process proceeds to step S20, and when there is no corresponding record, the process proceeds to step 26.

  In step S20, a keyword search is performed on the character string in the text of the content “n”. In this step S20, text search is performed within a single record without combining the preceding and succeeding records. For example, in the example of FIG. 3, a keyword search is performed on the character string in the content ID “2”.

  As a result, if the corresponding character string is not extracted, a keyword search is also performed on the character string in the text of the subsequent record ID “n + 1” (S22a). On the other hand, when the corresponding character string (ie, search keyword) is extracted in the single record “n”, the content ID information and record ID information of the record ID “n” are added to the search list (S22b). . That is, in this case, a search keyword exists in the record ID “n”.

  If the corresponding character string (ie, search keyword) is not extracted even in the single record “n + 1”, the content ID information and record ID information of both record IDs “n” and “n + 1” are stored in the search list. It is added (S24a). That is, in this case, the search keyword exists across the record IDs “n” and “n + 1”.

  On the other hand, if the corresponding character string (ie, search keyword) is extracted in the single record “n + 1”, the content ID information and record ID information of the record ID “n + 1” are added to the search list (S24b). . That is, in this case, a search keyword exists in the record ID “n + 1”.

  Subsequently, the next content IDs “n + 1” and “n + 2” are similarly searched (S15 to S25). Then, a search is performed for all record IDs belonging to one content.

  Next, from the search result list obtained so far, the data with the same content ID and the same item in the content basic information table is acquired (S28). Then, the acquired data is converted into display data (S29). In the present invention, the aspect of the display data is not particularly limited, but may be an aspect listed on the user terminal, for example.

  Subsequently, data in which the corresponding content ID matches the same item in the content basic information table is acquired (S30). Then, the acquired data is converted into display data (S31). In the present invention, the aspect of the display data is not particularly limited, but may be an aspect listed on the user terminal, for example. Then, Steps S28 to S31 are performed in the same manner for all the items in the corresponding search list (S27 to 32).

  FIG. 6 is a flowchart when outputting the detection result in the first embodiment. That is, FIG. 6 is a flowchart for outputting the search result obtained by the search processing in FIG. 5, and is a search result output for reproduction from a reproduction portion including a predetermined keyword.

  First, it is selected whether or not to output the playback position in the accuracy improvement mode (S33). When the accuracy improvement mode is selected, more accurate keyword playback from the playback position is possible. When the accuracy improvement mode is selected, first, the start time position of the corresponding content data is acquired from the result list obtained by the search process (see FIG. 4 and the like) (S34). Subsequently, either the Japanese or English audio information file to be output is selected (S35). In the present invention, the speech language is not limited to Japanese and English, and may be a plurality of types of languages.

  In the case of Japanese, noise cut processing based on Japanese speech is performed (S36a). In the case of English, noise cut processing based on English speech is performed (S36b). Since the number of utterance waves varies depending on the language, it can be reproduced with little noise when it is reproduced and output by performing an appropriate noise cut process according to the language. It is desirable to process this noise cut while reproducing the audio information file.

  Next, information on the start time position of the corresponding content data is acquired from the search result list obtained in the search process (see FIG. 4 and the like) (S36). By this step, information on the scheduled reproduction position of which audio ID file (that is, which content ID) should be reproduced from what record ID of what record ID is acquired. Then, according to the information of the scheduled reproduction position, reproduction is performed from the corresponding reproduction scheduled position (S37), and the reproduced part is converted into text (S38). Once the voice element of the voice information is converted to text and it is verified whether the text really includes the keyword character, the search accuracy can be improved.

  Then, it is determined whether or not the first character of the keyword is included in the text at the scheduled playback position (S39). For example, when a voice information file including the keyword “Internet” is searched, it is a step of determining whether or not the first character “I” of “Internet” is included in the scheduled playback position. As a result, if the first character is included, a reproduction position obtained by subtracting a predetermined time (one second) from the reproduction start time (scheduled reproduction position) of the first character is stored in a variable (S40).

  In this step 40, for example, if the keyword “Internet” is searched for and output for the content “create homepage” in FIG. 3, the record ID including “Internet” is “4”, and the keyword “Internet” When the playback start time at which the first character “I” of “Internet” is located is 35 seconds after the beginning of the content, the step is to store 34 seconds one second before the position as a playback position in a variable. is there. As a result, it is possible to prevent the voice information file from being played when the first character of “Internet” is cut off and the head is cut off like “Internet”, and the voice is played back right before “Internet”. Can do.

  In the present invention, the predetermined time does not have to be 1 second, and can be appropriately determined according to the type of the search target file, the audio content, and the like.

  If it is determined in step S39 that the first character of the keyword does not match, the voice information file at the scheduled start position is reproduced again (S37), and the second character of the keyword (n = 2). (S44).

  In this way, the same steps are performed for the second and subsequent characters of the keyword, and it is determined whether the keyword word (or phrase or sentence) is included in the text at the reproduction planned position (S41 to S45). That is, the determination of the second and subsequent characters of the keyword is similarly performed by reproducing the voice information file at the scheduled start position (S42) and converting it to text (S43). Then, it is determined whether the converted text matches the nth character (n = 3, 4,...) Of the keyword (S44). This loop of steps S41 to S45 is performed until all n characters of the keyword match.

  Then, the start time position of the corresponding content data in the search result list and the time confirmed in the variable are calculated (S46). For example, in the case of FIG. 3, when the reproduction start position of the keyword “Internet” is 35 seconds after the beginning of the content, the position of 34 seconds is calculated by subtracting 1 second which is a predetermined time from 35 seconds.

  In this way, the reproduction start position calculated in step S46 is set (designated) (S47), and reproduction is performed from the reproduction designated position of the predetermined audio information file (S48).

  If the accuracy improvement mode is not selected in step S33, the start time position of the corresponding content data is obtained as it is from the result list obtained by the search process (S10b), and is directly used from the start time position. Playback is performed (S47). For example, in the case of searching for the keyword “homepage” in the content data “homepage creation” in FIG. 3, the first playback start position of the record ID “4” containing the keyword (that is, 30 seconds from the beginning of the content). Play).

  That is, in the present invention, by providing the accuracy improvement mode, the accuracy improvement mode can be appropriately selected as “valid” or “invalid”. When the accuracy improvement mode is to function, the accuracy improvement mode is set to “valid”. As a result, reproduction is performed from the time stamp of the record to be searched, and the location matching the keyword is specified. Then, the total time of the time until matching and the time of the time stamp is set as the reproduction start position for the user.

  If the accuracy improvement mode is not desired to function, the accuracy improvement mode is set to “invalid”. As a result, reproduction is performed from the time stamp of the record to be searched (that is, from the beginning of the record ID to be searched).

  FIG. 7 is a flowchart showing a procedure for registering a database in the second embodiment of the voice information search program according to the present invention. In the second embodiment, the audio information file according to the present invention is registered in a database based on a predetermined number of characters. Hereinafter, the difference from the first embodiment of the present invention will be mainly described.

  First, the file information of the audio file is input (S1), and the procedure up to text conversion of the audio file (S6) is the same as in FIG. It is determined whether the number of characters in the voice information file converted into text in this way is larger or smaller than a predetermined number of characters (n characters in FIG. 7) (S7-2).

  As a result, if it is determined that the number is less than n, the text information of the voice information file is registered in the database (S8a). That is, the audio information file processed in step S8a is an audio information file composed of short text whose reproduction time is n characters or less.

  As a result of the determination in step S7-2, if it is determined that the number of characters is longer than n characters, the first n number of reproduction parts of the audio information file are separated, and the text information of the separated audio information file is stored in the database. Register (S8b). Then, with respect to the remaining portion of the cut voice information file, it is determined again in step S7-2 whether the text is longer than the number of n characters, and finally the voice information file is divided in units of n characters. . That is, the audio information file processed in step S8b is an audio information file in which the number of texts in the reproduction information is longer than a predetermined number of characters (n characters).

  That is, in the present invention, the reproduced audio information is converted into text and stored in the database. At that time, by pressing a time stamp in a repeating unit of a predetermined number of characters (number of n characters), it is divided into units. To be recorded. If the audio information has a short information amount equal to or less than the predetermined number of characters (n characters), the time stamp is pressed and recorded as it is without being divided.

  Thus, by dividing the voice information by a predetermined number of characters, it is possible to cut the unvoiced part or the music-only part in the voice information. In other words, since no record is created for the silent portion or the like, data enlargement can be prevented. Therefore, even for a long-time file, text data (data obtained by converting voice information into text) can be stored with a minimum data capacity. This method is particularly effective when a long-time audio file including many parts other than lyrics and lines such as music, movies, or musicals is targeted.

  The search process and the search result output process after the audio information file is registered in the database can be performed in the same procedure as in the first embodiment of the present invention (see FIGS. 5 and 6). .

  FIG. 8 is a flowchart showing a procedure for registering a database in the third embodiment of the speech information search program according to the present invention. In the third embodiment, the audio information file in the present invention is registered in a database based on a predetermined number of words. The following description will focus on differences from the first and second embodiments of the present invention.

  First, the file information of the audio information file is input (S1), and the procedure up to text conversion of the audio information file (S6) is the same as that shown in FIGS. The voice information file converted into text in this way is divided into clauses or words as a unit (S7-3).

  That is, in the present embodiment, the text is divided into predetermined clause units or word units, and the unit for the division may be a phrase unit or a word unit, and can be selected as appropriate. And in this invention, it does not specifically limit about the method of recognizing the said phrase unit or a word unit, A suitable method can be used suitably. Also, the unit for dividing is not particularly limited. For example, although not illustrated, a predetermined phrase unit (or word unit) can be set as the number of n clauses (or n words), and the unit for dividing is shortened. The search accuracy is increased, and if the length is increased, the search speed can be increased, and a suitable unit can be appropriately determined in consideration of these.

  As a result, when the number of phrases (or the number of words) of the voice information file subjected to text conversion is equal to or less than the predetermined n number of phrases (or the number of words), it is registered in the database at the end of reproduction and the processing is ended ( S8a). That is, the audio information file processed in step S8a is an audio information file composed of short text whose reproduction time is the number of possessed phrases n (predetermined number of words n) or less.

  As a result of the determination in step S7-3, if it is determined that the number is longer than the number of n clauses (n words), the playback part of the first n phrases (n words) of the audio information file is divided, Text information of the divided audio information file is registered in the database (S8b). Then, with respect to the remaining portion of the cut voice information file, it is determined again in step S7-3 whether or not the text is longer than n clauses (n words), and finally the voice information file becomes n. It is delimited by the number of phrases (n words). That is, the audio information file processed in step S8b is an audio information file in which the number of texts in the reproduction information is longer than the predetermined number of phrases (predetermined number of words).

  FIG. 9 is a flowchart for performing a search process on the database according to the third embodiment. That is, FIG. 9 is a flowchart for performing the search process for the database registered in FIG. Hereinafter, differences from the case where the audio information file is registered in the database based on a predetermined time (see FIG. 6 and the like) will be described.

  First, the search condition is input (S9), and the procedure (S10, S11) for extracting data from the content basic information table from the predetermined condition such as category / language is the same as that of the first and second embodiments of the present invention. It is the same.

  If the number of extracted items is one or more in step S11, a keyword in the text of the content data table is searched from the content ID of the corresponding data (S49a). That is, it is a step of further searching the content ID.

  If the number of extractions is 0 in the search result of step S11, “N / A” is displayed and the search ends (S49b).

  If there is one or more extraction results as a result of keyword search in step S49a, a loop search is performed for the corresponding number (S51a to S57). In step S49a, if the number of extractions is 0, “not applicable” is displayed and the search is terminated (S49b).

  When the number of extractions is one or more, the following loop search is performed (S51a to S57). First, the content ID and the record ID “n” are added to the result list. That is, by this step, the record ID including the corresponding voice part is specified. Further, data in which the corresponding content ID matches the same item in the content basic information table is acquired from the result list (S53). Subsequently, the acquired data is converted into display data (S54). For example, it may be display data that can be displayed as a list on the user's terminal device and the corresponding portion can be reproduced as it is by clicking on the linked portion.

  A loop search is performed in which the series of steps (S52 to S56) is performed for all the corresponding cases (S57).

  Thus, as in the third embodiment according to the present invention, by separating the audio information based on the number of phrases and the number of words, the search is performed by combining the adjacent record ID “n” and the record ID “n + 1”. High-accuracy voice information retrieval can be performed without performing the processing (for example, see S15 to S25 in FIG. 5) or the like. That is, since it is registered in the database in a phrase unit or word unit that makes sense in advance, the record ID ends in a truncated state (for example, see record ID “2” of content ID “1” in FIG. 4 etc.) ) Can be prevented.

  In this way, by registering and processing based on the number of phrases (or the number of words) as in this embodiment, it is possible to register in the database by omitting portions such as unvoiced parts in the audio file, and record of text data Since there is no need to search by combining IDs, the load on the server can be reduced, and a higher speed search can be performed.

  FIG. 10 is a flowchart when outputting the detection result in the third embodiment. That is, FIG. 10 is a flowchart for outputting the search result obtained by the search process in FIG. Hereinafter, differences from the case where the audio information file is registered in the database based on a predetermined time (see FIG. 7 and the like) will be described.

  The start time position of the corresponding content data is acquired from the result list obtained by the search process performed in FIG. 9 (S58). Subsequently, the start time of the content data is set (S59). Then, reproduction is performed from the set start time (S60).

  Since registration / processing is performed based on the number of phrases (or the number of words) as in the present embodiment, data can be registered in phrase units (or word units). There is no need to set / select the accuracy improvement mode (see S33 in FIG. 6) at the time of retrieval as in the second embodiment. Therefore, faster search processing can be performed, the load on the server can be further reduced, and a search with high search accuracy similar to the accuracy improvement mode can be performed. Furthermore, since it is possible to deal with a wide variety of audio information files (for example, moving image files) and the burden on the server is small, a larger amount of data can be handled widely.

  As described above, in the present invention, when the voice information file is converted into text, it may be divided by a predetermined time unit (see FIG. 2, etc.), may be divided by a predetermined number of characters (see FIG. 7, etc.), or predetermined It may be divided by the number of phrases or the number of words (see FIG. 8 etc.). These should be selected as appropriate in consideration of the processing capability when converting an audio information file into text and the processing capability such as whether the text-converted file can be recognized as a character or a phrase or a word. Can do.

  As described above, in the present invention, by registering the voice information file as text data in a certain unit in the database, etc., it is possible to perform voice information search with high accuracy while reducing the burden on the server while maintaining a high search speed. . Therefore, the present invention can be used widely and easily for various databases.

  FIG. 11 is a conceptual diagram for explaining an example of the voice information retrieval system according to the present invention.

  The speech information retrieval system according to the present invention is a system in which a text conversion server 8, a database server 9, a Web server 10, a speech information file storage server 11, and a management server 12 are connected via a network 13.

  The text conversion server 8 is a server that converts a voice information file into text, creates a content information file, and registers it in a database. The database server 9 is a server that stores the text-converted information file 81 and the content information file 82. The Web server 10 is a server that is connected to the user terminal 91 and exhibits a search condition input from the user terminal 91 and a function of displaying and outputting the search result. The audio information file storage server 11 is a server that stores audio information files (including moving image information files). The management server 12 is a server that manages system failures such as each server and the network 13.

  Further, the accuracy improvement mode can be selected as necessary. When the accuracy improvement mode is enabled, the accuracy improvement mode is set to “valid”. As a result, reproduction is performed from the time stamp of the record to be searched, and the location matching the keyword is specified. Then, the total time of the time until matching and the time of the time stamp is set as the reproduction start position for the user.

  If the accuracy improvement mode is not desired to function, the accuracy improvement mode is set to “invalid”. As a result, reproduction is performed from the time stamp of the record to be searched (that is, from the beginning of the record to be searched).

  In the present invention, the ability of the database can be adjusted by shortening the time required for search or improving the search accuracy by adjusting settings as appropriate. For example, by setting the accuracy improvement mode to be effective, the search accuracy of voice information is improved, the position where a user's desired keyword is output (spoken) can be detected, and a more accurate playback start position is searched. it can. On the other hand, by disabling the accuracy improvement mode, the step of the accuracy improvement mode is unnecessary, so that the time required for search can be shortened and the load on the server of the search system can be reduced. That is, according to the present invention, it is possible to appropriately construct a search system in consideration of the type and amount of information file to be searched, or the hardware environment to be used.

  Furthermore, the information recorded in the database used in the present invention is not limited to the audio information file (or moving image information file), and for example, a text information file can also be recorded. In the present invention, since voice information is converted into text and registered in the database, not only the voice information file but also the text information file can be searched simultaneously. That is, in the present invention, not only the audio information file but also the text information file is registered, so that the audio information file (or moving picture information file) and the text information file are simultaneously searched by executing one search. be able to.

  In the present invention, a step for designating a search target may be separately provided before executing the search. Thereby, it is possible to make an appropriate selection in consideration of the user's purpose of use, usage environment, and the like, such as a search for only an audio information file or a search for only a text information file.

  And in this invention, when performing a search process, you may provide the function which can search a some content simultaneously. By performing a simultaneous search for a plurality of contents, the time required for the search process can be shortened. In this case, it is possible to appropriately determine whether to provide this function in consideration of the processing capability of the computer or server used in the present invention.

  The audio information search program according to the present invention can be used for management, aggregation, and search of various audio information files, and in particular, a multimedia-related database for managing audio information files with a large amount of information, and a huge amount of information files The number can be incorporated into a search engine on the Internet or the like for which the number is to be searched.

It is a conceptual diagram for demonstrating the audio | voice information retrieval program which concerns on this invention. It is a flowchart at the time of registering the database in 1st Embodiment of the audio | voice information program which concerns on this invention. In the same 1st Embodiment, it is a conceptual diagram explaining the basic information table of the audio | voice information file registered into the database. In the same 1st Embodiment, it is a conceptual diagram explaining the content data of the audio | voice information file registered into the database. It is a flowchart at the time of searching for the database in the first embodiment. It is a flowchart at the time of outputting the detection result in the first embodiment. It is a flowchart at the time of registering the database in 2nd Embodiment of the audio | voice information program which concerns on this invention. It is a flowchart at the time of registering the database in 3rd Embodiment of the audio | voice information program which concerns on this invention. It is a flowchart at the time of searching about the database in the same 3rd embodiment. It is a flowchart at the time of outputting the detection result in the said 3rd Embodiment. It is a conceptual diagram for demonstrating an example of the audio | voice information retrieval system which concerns on this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Text conversion function of voice information file 2 Playback time measurement function of voice information file 3 Registration function to database 4 Search condition determination function 5 Search processing function 6 Search result output function 7 Database 8 Text conversion server 9 Database server 10 Web server 11 Voice information file storage server 12 Management server 13 Network

Claims (9)

Converting a voice element recorded in the voice information into a character string;
Measuring the playback time of the audio information;
When the reproduction time of the audio information is longer than a predetermined time, the character string of the audio information is registered in a database divided every predetermined time, and when the reproduction time of the audio information is less than the predetermined time, Registering the voice information in the database without dividing the character string;
Determining a search item for searching the database, and a search condition for the search item;
Performing a search on the database according to the search condition;
Outputting the result of the search;
A speech information retrieval program for causing a computer to execute at least. Converting a voice element recorded in the voice information into a character string;
Measuring the number of characters of the voice information;
When the number of characters in the character string is larger than a predetermined number of characters, the characters constituting the character string are divided into predetermined characters and registered in a database, and when the number of characters in the character string is equal to or less than the predetermined number of characters, Registering in the database without separating the strings;
Determining a search item for searching the database, and a search condition for the search item;
Performing a search on the database according to the search condition;
Outputting the result of the search;
A speech information retrieval program for causing a computer to execute at least. The voice information search program according to claim 1 or 2, wherein the computer is caused to execute at least the following steps in order to execute a search on the database.
(1) A step of combining two consecutive pieces of audio information before and after the divided audio information.
(2) A step of determining whether or not the character string specified by the search condition is included in the character string of the two combined voice information.
(3) A step of determining whether the character string specified by the search condition is included in the character string of each voice information in the character strings of the two divided voice information. 3. The speech information search program according to claim 1, wherein the computer executes at least the following steps in order to output the search result. 4.
(A) A step of designating a recording time position of voice information in which a voice element corresponding to the character designated by the search condition is recorded.
(B) A step of designating a recording time position that goes back a predetermined time before the designated recording time position as a reproduction start position.
(C) Step of reproducing sound from the designated reproduction start position   A computer-readable recording medium in which the voice information retrieval program according to any one of claims 1 to 4 is recorded. Means for converting a voice element recorded in voice information into a character string;
Means for measuring the playback time of the audio information;
When the reproduction time of the audio information is longer than a predetermined time, the character string of the audio information is registered in a database divided every predetermined time, and when the reproduction time of the audio information is less than the predetermined time, Means for registering the voice information in the database without dividing the character string;
A search item for performing a search on the database; a search condition determining means for the search item;
Means for performing a search on the database according to the search condition;
Means for outputting the search results;
Voice information retrieval system comprising at least Means for converting a voice element recorded in voice information into a character string;
A means for measuring the number of characters of the voice information;
When the number of characters in the character string is larger than a predetermined number of characters, the characters constituting the character string are divided into predetermined characters and registered in a database, and when the number of characters in the character string is equal to or less than the predetermined number of characters, A means of registering in the database without separating strings,
Means for determining a search item for searching the database, and a search condition for the search item;
Means for performing a search on the database according to the search condition;
Means for outputting the search result;
Voice information retrieval system comprising at least A procedure for converting a voice element recorded in voice information into a character string;
Measuring the playback time of the audio information;
When the reproduction time of the audio information is longer than a predetermined time, the character string of the audio information is registered in a database divided every predetermined time, and when the reproduction time of the audio information is less than the predetermined time, A procedure for registering the voice information in the database without dividing the character string;
A procedure for determining a search item for searching the database and a search condition for the search item;
Performing a search on the database according to the search condition;
Outputting the search results;
Voice information retrieval method comprising at least A procedure for converting a voice element recorded in voice information into a character string;
A procedure for measuring the number of characters of the voice information;
When the number of characters in the character string is larger than a predetermined number of characters, the characters constituting the character string are divided into predetermined characters and registered in a database, and when the number of characters in the character string is equal to or less than the predetermined number of characters, The procedure to register in the database without separating the character string,
A procedure for determining a search item for searching the database and a search condition for the search item;
Performing a search on the database according to the search condition;
Outputting the search results;
Voice information retrieval method comprising at least