JP2004295102A5 - - Google Patents

Description

Voice recognition dictionary creation device and information retrieval device

  The present invention relates to a speech recognition dictionary creation device and an information search device for creating a speech recognition dictionary for performing speech recognition by an information search device such as a program guide device.

  2. Description of the Related Art Conventionally, a program guide device using a voice recognition dictionary creating device is a device that performs TV operation using an electronic program guide (EPG) and voice along with an increase in the number of channels of a program. After receiving the data and using the received data to create a speech recognition dictionary corresponding to the program of today or tomorrow or later, the old recognition dictionary is erased to avoid using extra hardware resources.

  Further, when a dictionary is created, the dictionary is classified into “program name”, “genre”, “performers”, etc., which are categorized in the EPG, and the dictionary is registered. Each word is recognized equally.

  FIG. 28 is a block diagram showing a configuration of a program guide device provided with a conventional voice recognition dictionary creating device.

The program guide device includes an EPG receiving unit 701, an item classifying unit 702, a dictionary updating unit 703, a dictionary 704, a voice recognition unit 705, a clock 706, a registered word deleting unit 707, an EPG storage unit 708, and a control unit 709. .
EPG receiving section 701 separates EPG data from a television signal. The item classification unit 702 separates EPG data such as “broadcast time” that is not necessary for speech recognition, and classifies the remaining EPG data into “program titles”, “performers”, and the like. The dictionary updating unit 703 registers words included in the EPG data in the dictionary without duplication. The dictionary 704 stores words necessary for speech recognition together with the registration date and time. The speech recognition unit 705 refers to the dictionary 704 and recognizes a specific word from a speech signal generated by the user's speech.

The clock 706 outputs the current date and time to the registered word deletion unit 707. The registered word deletion unit 707 refers to the date and time of the clock 706 and deletes, for example, words registered up to the previous day among words registered in the dictionary 704. The EPG storage unit 708 stores the EPG data separated by the EPG receiving unit 701. The control unit 709 searches the EPG data stored in the EPG storage unit 708 for a word recognized by the voice recognition unit 705, and outputs the corresponding program content (for example, see Patent Document 1).
JP 2001-22374 A (Page 7, FIG. 1)

  In the conventional program guide device as described above, since the dictionary is created using the latest EPG data, the recognition word in the dictionary changes, but the user is not so conscious of the change in the recognition word in the dictionary. .

  Here, for example, when the user performs a TV program search and operation using a program name or a performer name that was interesting in a past-viewed program that is not included in the latest EPG data, the program guide device uses The word having the highest recognition score is output as a recognition result. This is because the program guide device cannot recognize words that are not registered in the dictionary. For this reason, the output recognition result may be different from the program name and cast name requested by the user. In this case, the user considers this recognition result to be erroneous recognition and utters the same word many times.

  Therefore, it is conceivable to create a dictionary using not only the latest EPG data but also past EPG data. In this case, however, the number of words registered in the dictionary is enormous.

  Further, as the number of programs increases due to the increase in the number of channels, the number of words to be recognized also increases. However, since the conventional speech recognition dictionary creation device uses new data as all the words to be recognized, there is a problem that the recognition rate is reduced.

  Therefore, the present invention has been made in view of the above circumstances, and provides a speech recognition dictionary creating device and an information search device capable of improving the recognition rate while minimizing the increase in recognition target words. With the goal.

In order to achieve the above object, a speech recognition dictionary creation device according to the present invention is a speech recognition dictionary creation device for creating a recognition dictionary for performing speech recognition on data relating to program information or music content, and A viewing history word acquiring unit that acquires a viewing history word that is a word to be recognized based on a viewing experience related to the data, a viewing history word accumulating unit that accumulates the acquired viewing history word, and a predetermined predetermined Fixed word accumulating means for accumulating fixed words, word extracting means for extracting a new recognition word that is a word to be recognized from input program information or data relating to music content, and the new recognition word Dictionary creating means for creating a plurality of the recognition dictionaries having different contents from the fixed word and the viewing history word; And characterized in that it comprises a dictionary storing means for storing said recognition dictionary.

Further, an information search device according to the present invention is an information search device that creates a recognition dictionary for performing voice recognition on program information or data relating to music content, and searches for information using a voice signal generated by a user's speech. A viewing history word acquiring unit that acquires a viewing history word that is a word to be recognized based on the viewing experience of the user regarding the data, a viewing history word storing unit that stores the acquired viewing history word, Fixed word storage means for storing a predetermined fixed word to be set, word extraction means for extracting a new recognition word that is a word to be recognized from input program information or data relating to music content, A word for creating a plurality of the recognition dictionaries having different contents from the new recognition word, the fixed word, and the viewing history word. Creation means, dictionary storage means for storing the created recognition dictionary, voice recognition means for recognizing a specific word from the voice signal by referring to the recognition dictionary, and data for storing input data It is characterized by comprising a storage unit, and a control unit that searches the data for a word recognized by the voice recognition unit from the data and outputs corresponding information.

  As described above, according to the speech recognition dictionary creating apparatus of the present invention, words that are not included in the latest EPG data can be treated as words to be recognized, and when a program search is performed, a speech recognition error And the presence or absence of search data, a user-friendly voice interface can be realized.

  Furthermore, by creating a plurality of recognition dictionaries and switching from the designation of the user, the user who thinks that the recognition rate is low can select the dictionary that is considered optimal from the plurality of dictionaries, thereby increasing the recognition rate. Improved and easy-to-use interface can be realized.

An apparatus for creating a speech recognition dictionary according to an embodiment of the present invention is a speech recognition dictionary creating apparatus for creating a recognition dictionary used for speech recognition, wherein the content of the recognition differs depending on whether the user has experience. It is characterized by comprising dictionary creation means for creating a plurality of dictionaries.
Here, the presence or absence of the experience is the presence or absence of the viewing experience, and the dictionary creating unit may create a plurality of the recognition dictionaries having different contents according to the presence or absence of the viewing experience.

Further, the presence / absence of the experience is the presence / absence of a use experience, and the dictionary creating unit may create a plurality of the recognition dictionaries having different contents according to the presence / absence of the use experience.
Further, the presence or absence of the experience may be presence or absence of a visit experience, and the dictionary creating unit may create a plurality of the recognition dictionaries having different contents according to the presence or absence of the visit experience.
Thereby, for example, a plurality of recognition dictionaries are determined according to whether or not the user has a viewing experience, such as whether to search for a program to be broadcasted in the future or a program that has been watched (or recorded) before. Since it is created, it is possible to create a recognition dictionary with an improved recognition rate of speech recognition according to the purpose.

  Further, the voice recognition dictionary creation device further includes: an action history word acquiring unit that acquires an action history word that is a word to be recognized based on a user's action; and an action that accumulates the acquired action history word. History word accumulating means, fixed word accumulating means for accumulating a predetermined fixed word set in advance, word extracting means for extracting a new recognition word which is a word to be recognized from input data, creation Dictionary storage means for storing the recognized recognition dictionary, wherein the dictionary creation means stores a new registration word that is a word to be registered in the recognition dictionary from the new recognition word, the fixed word, and the action history word. The recognition dictionary may be created by determining and adding information indicating the reading kana and the part of speech of the newly registered word to the newly registered word.

  Thus, by registering action history words that are not included in the input data, it is possible to create a recognition dictionary that suppresses erroneous speech recognition.

Further, the speech recognition dictionary creation device further includes a recognition word acquisition unit that acquires a past recognition word that is a word recognized by speech recognition, and a past recognition word accumulation unit that accumulates the past recognition word. The dictionary creating means determines a new registration word to be registered in a recognition dictionary from the new recognition word, the fixed word, the action history word, and the past recognition word, and determines the new registration word as the new registration word. The recognition dictionary may be created by adding information indicating the reading kana and the part of speech of the registered word.
Thus, by registering past recognition words that are not included in the input data, it is possible to create a recognition dictionary that suppresses erroneous recognition of voice recognition.

  In this way, instead of registering all words that existed in the past certain period as recognition targets in addition to the latest words, only the action history words or words that are past recognition words are registered, so the recognition target words Is suppressed to a minimum, and the recognition rate can be reduced with a small amount of hardware resources.

Further, the dictionary creating means weights the newly registered word based on whether or not the word is included in the action history word, and creates the recognition dictionary by adding information relating to the weighting. Is also good.
This makes it possible to create a recognition dictionary in which the recognition rate of speech recognition is improved by weighting newly registered words based on whether or not the words are included in the action history words.

Further, the dictionary creating means weights the newly registered word based on whether or not the word is included in the past recognized words, and creates the recognition dictionary by adding information on the weighting. Is also good.
This makes it possible to create a recognition dictionary in which the recognition rate of speech recognition is improved by weighting newly registered words based on whether or not the words are included in past recognized words.

Further, the input data is structured data, the word extracting means extracts the attribute of the new recognition word together with the new recognition word, and the dictionary creating means outputs the new recognition word. The newly registered word may be weighted based on a word attribute, and the recognition dictionary may be created by adding information on the weight.
This makes it possible to create a recognition dictionary in which the recognition rate of speech recognition is improved by weighting newly registered words based on the attributes of the words for new recognition.

Further, the voice recognition dictionary creation device further includes user information acquisition means for acquiring user recognition information for identifying the user, wherein the dictionary creation means stores the recognition dictionary for each of the user recognition information. May be created.
Further, the voice recognition dictionary creation device further includes user information acquisition means for acquiring user recognition information for identifying a user, wherein the dictionary creation means includes the new registration word for each of the user recognition information. The recognition dictionary may be created by performing weighting on the recognition dictionary.
As a result, a recognition dictionary suitable for a plurality of users can be created.

  Further, the action history word acquiring means records the action date and time together with the action history word in the action history word storage means, and the dictionary creation means further includes the action date and time in the action history word among the newly registered words. To generate the recognition dictionary, and the speech recognition dictionary creating device further includes a word deletion unit that deletes the registered word for which a predetermined period has elapsed from the action date and time of the registered word from the recognition dictionary. You may.

  Thus, by using a recognition dictionary having words that are not included in the input data, erroneous recognition of speech recognition is suppressed, and registered words are deleted from the recognition dictionary after a predetermined period has elapsed. Therefore, the hardware resources of the speech recognition dictionary creation device can be reduced.

Further, an information search device according to an embodiment of the present invention is an information search device that creates a recognition dictionary used for voice recognition and searches for information by a voice signal generated by a user's speech. Dictionary creation means for creating a plurality of the recognition dictionaries having different contents according to the presence or absence of the speech recognition means, and speech recognition means for recognizing a specific word from the speech signal by referring to the recognition dictionary. Data storage means for storing data, and control means for searching a word recognized by the voice recognition means from the data and outputting corresponding information.

  The information search device may further include an action history word acquiring unit that acquires an action history word that is a word to be recognized based on a user's action, and an action history word that accumulates the acquired action history word. A storage unit, a fixed word storage unit that stores a predetermined fixed word set in advance, a word extraction unit that extracts a new recognition word that is a word to be recognized from input data, Dictionary storage means for storing the recognition dictionary, wherein the dictionary creation means determines a new registration word that is a word to be registered in the recognition dictionary from the new recognition word, the fixed word, and the action history word. The recognition dictionary may be created by adding information indicating the reading kana and the part of speech of the newly registered word to the newly registered word.

  Thus, by using a recognition dictionary having words that are not included in the input data, erroneous recognition of speech recognition can be suppressed, and it is possible to identify erroneous speech recognition and the presence or absence of search data.

  For example, if the information retrieval device is used as a program guide device for a television broadcast program, words can be correctly recognized even if the user utters a previously viewed program that is not included in the latest EPG data. As a result, when a program search is performed, it is possible to reply that "The program in which Talent C appears is not currently being broadcast and within one week" and "Movie K is not currently being broadcast". Thus, a user-friendly voice interface can be realized.

  Further, the input data is structured data, the word extracting means extracts the attribute of the new recognition word together with the new recognition word, and the dictionary creating means outputs the new recognition word. The newly registered word is weighted based on the attribute of the word, the information on the weight is added to create the recognition dictionary, and the new registered word based on the attribute of the new recognition word is added according to the purpose. A plurality of the recognition dictionaries having different weights are created, and the information search device further includes a dictionary selection receiving unit that receives a dictionary selected by a user from the plurality of recognition dictionaries stored in the dictionary storage unit. The voice recognition unit refers to a recognition dictionary received by the dictionary selection reception unit, and thereby specially converts a voice signal generated by a user's utterance. Words may recognize.

  Thus, a user who thinks that the recognition rate of speech recognition is low selects a dictionary that is considered optimal from a plurality of dictionaries, thereby improving the recognition rate of speech recognition and realizing an easy-to-use interface.

  Further, the present invention can be realized not only as such a device, but also as a method including, as steps, characteristic means included in such a speech recognition dictionary creating device, or executing those steps on a computer. Or as a program that causes Needless to say, such a program can be distributed via a recording medium such as a CD-ROM or a transmission medium such as the Internet.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
(Embodiment 1)
In the first embodiment, a case will be mainly described in which a user creates a recognition dictionary for performing a search for a program to be broadcasted in the future and performs a search.
FIG. 1 is a block diagram showing a configuration of a program guide device including a speech recognition dictionary creation device according to Embodiment 1 of the present invention.

The program guide device includes an EPG receiving unit 101, a word extracting unit 102, a dictionary creating unit 103, a fixed word storing unit 104, a recognized word storing unit 105, a viewing history word storing unit 106, a dictionary storing unit 107, a dictionary selecting unit 108, and a voice. A recognition unit 109, an EPG storage unit 110, a clock 111, and a control unit 112 are provided.
EPG receiving section 101 separates EPG data from the input television signal. The word extracting unit 102 extracts a new recognition word that is a recognition target word such as “program title”, “performer”, or “genre” from the EPG data. The fixed word storage unit 104 stores words indicating time and broadcasting stations and words necessary for normal speech recognition, for example, "morning", "6:00", "no", "news", "", " Fixed words such as "tai" are accumulated.

  The recognized word storage unit 105 is a storage unit that stores past recognized words that are words recognized in the past by voice recognition. Here, for example, as shown in FIG. 2, "notation" indicating the notation of the word recognized by speech, "yomi" indicating the reading kana of the word, and the word may be assigned to any of the program name, genre, cast, etc. “Type” indicating whether the message is applicable, “Speech recognition date and time” indicating the date and time of voice recognition, and “frequency” indicating the number of times of voice recognition are recorded. When words having the same notation are recognized, the “speech recognition date” is updated with the latest date and the “frequency” is increased by one.

  The viewing history word storage unit 106 is a storage unit that stores viewing history words, which are information on viewed programs. Here, the program name and the performer are recorded as information of the viewed program, for example, as shown in FIG. 3, “notation”, “viewing date and time” indicating the date and time when the program was viewed, “type”, and “frequency” Is recorded. When words having the same notation are recorded, the “viewing date and time” is updated with the latest date and time, and the “frequency” is increased by one.

  The dictionary creation unit 103 adds information such as a reading kana and a part of speech necessary for speech recognition to the new recognition word extracted by the word extraction unit 102 and the fixed words stored in the fixed word storage unit 104. For example, a basic recognition dictionary as shown in FIG. Further, the dictionary creating unit 103 refers to the past recognized words stored in the recognized word storage unit 105 and the viewing history words stored in the viewing history word storage unit 106, and refers to the words other than the words registered in the basic recognition dictionary. If the word exists, the word is added to the word registered in the basic recognition dictionary to create a user adaptive recognition dictionary as shown in FIG. 5, for example.

  By the way, the type (attribute) of a word uttered by the user for searching using the program guide device is different from the case where the user searches for a program to be broadcasted in the future (utterance condition 1). There is a difference in the utterance frequency between the case where a search is performed for a program that has been watched or recorded (utterance condition 2).

  FIG. 6 is a diagram illustrating a result of investigating a frequency at which a type of word is uttered under each utterance condition. As shown in FIG. 6, excluding the word of “date and time”, in the case of the utterance condition 1, “genre” (44.8%) is the largest, followed by “program content” (9.5%). On the other hand, in the case of the utterance condition 2, "program name" (26.7%) is the largest, followed by "genre" (23.3%).

  For this reason, the dictionary creation unit 103 performs weighting according to the type of word and determines a “target weight coefficient”.

  Further, the dictionary creation unit 103 performs weighting according to whether the recognition word is stored in the recognition word storage unit 105 as a past recognition word and whether or not it is stored in the viewing history word storage unit 106 as a viewing history word. Is determined. Then, as shown in FIG. 7, for example, as shown in FIG. 7, the dictionary creation unit 103 stores a “target weight coefficient” and a “history weight coefficient” in addition to “notation”, “read”, and “type”. Create a recognition dictionary.

  Here, as the “target weighting factor”, “1.11” is used for a word whose “type” is a program name, “1.27” is used for a word whose “type” is a date and time, and “1. “1.06” is assigned to a word that is a channel, “1.49” is assigned to a word whose “type” is a genre, and “1.07” is assigned to a word whose “type” is a performer.

  Further, as the “history weighting factor”, “1.2” is used for a word stored in the recognized word storage unit 105 as a past recognized word, and “1.2” is used for a word stored as a viewing history word in the viewing history word storage unit 106. “1.1” is assigned to the word stored in both the recognized word storage unit 105 and the viewing history word storage unit 106, and “1.3” is assigned to the word.

  The dictionary storage unit 107 is a storage unit that stores a plurality of dictionaries created by the dictionary creation unit 103 and used for speech recognition. The dictionary selection unit 108 receives a dictionary selected from a plurality of dictionaries by the user as appropriate. The voice recognition unit 109 recognizes a specific word from a voice signal generated by the user by referring to the dictionary accepted by the dictionary selection unit 108. The clock 111 outputs the current date and time to the control unit 112. The EPG storage unit 110 is a storage unit that stores EPG data.

  The control unit 112 searches the EPG data stored in the EPG storage unit 110 based on the word recognized by the voice recognition unit 109, and outputs information on the corresponding program. Further, the control unit 112 outputs information on the word recognized by the voice recognition unit 109, voice recognition date and time, and the like to the recognized word storage unit 105. Further, when a user watches a program for a predetermined time or more, control unit 112 acquires a viewing history word, which is information on the program, and outputs it to viewing history word storage unit 106. Further, the control unit 112 deletes past recognized words for which a predetermined time has elapsed from the voice recognition date and time among the past recognized words stored in the recognized word storage unit 105. Similarly, the control unit 112 deletes a viewing history word having a predetermined time elapsed from the viewing date and time among the viewing history words stored in the viewing history word storage unit 106.

  The operation of the program guide apparatus configured as described above can be broadly divided into an offline processing step of creating a dictionary and an online processing step of recognizing input speech using the created dictionary and guiding a program.

  First, the operation of the offline processing step will be described. FIG. 8 is a flowchart showing the operation of the offline processing step in the program guide device.

The EPG receiving unit 101 to which the television signal has been input extracts the EPG data from the television signal and outputs it to the word extraction unit 102 and the EPG storage unit 110 (step S101).
Next, the word extracting unit 102 extracts “program name”, “performer”, and “genre” from the EPG data as new recognition words to be keywords for voice recognition (step S102).

  FIG. 9 is a diagram illustrating an example of the EPG data. The EPG data includes, for each program, a “service ID” for identifying a broadcast station, an “event ID” for identifying each program, a “start date” indicating the date when the program starts, “Start time” indicating the time at which the program starts, “duration” indicating the broadcast time of the program, “event name” indicating the program name, “text” indicating the performers and the program contents, and “genre of the program” Genre ".

  The fixed words related to the items of “service ID”, “event ID”, “start date”, “start time”, and “duration” are stored in the fixed word storage unit 104 in advance.

  Next, the dictionary creating unit 103 generates a new recognition word input from the word extraction unit 102, a fixed word stored in the fixed word storage unit 104, and a reading kana necessary for speech recognition stored in the fixed word storage unit 104. A basic recognition dictionary is created in the same manner as a conventional recognition dictionary using information such as part of speech (step S103).

  Next, the dictionary creation unit 103 refers to the viewing history word input from the viewing history word storage unit 106 and the past recognition word input from the recognition word storage unit 105, and determines whether a word other than the word registered in the basic recognition dictionary is used. If it exists, the word is added to the word registered in the basic recognition dictionary to create a user adaptive recognition dictionary as shown in FIG. 5, for example. Further, the dictionary creating unit 103 determines the weight according to the type of the word, and determines whether the word is stored as a past recognized word in the recognized word storage unit 105 and whether it is stored as a viewing history word in the viewing history word storage unit 106. A weighted user-adaptive recognition dictionary that stores, for example, as shown in FIG. 7, “target weight coefficient” and “history weight coefficient” in addition to “notation”, “read”, and “type” as shown in FIG. Is created (step S104).

  Next, the dictionary creation unit 103 stores the created basic recognition dictionary, the user adaptive recognition dictionary, and the weighted user adaptive recognition dictionary in the dictionary storage unit 107 (step S105). At this time, the dictionary creation unit 103 updates by replacing each of the previously created recognition dictionaries stored in the dictionary storage unit 107.

  In the present embodiment, a dictionary storing “target weight coefficients” and “history weight coefficients” is created as a weighted user-adaptive recognition dictionary, but the present invention is not limited to this. For example, a dictionary to which only one of “target weight coefficient” and “history weight coefficient” is added may be used. Further, the values of the “target weight coefficient” and the “history weight coefficient” are not limited to the above examples, and may be other values.

Further, the “history weighting coefficient” is weighted based on the past recognition word and the viewing history word, but it is also possible to weight only one of them. Further, a “weight coefficient” for each of the past recognition word and the viewing history word may be added to the dictionary. Further, the “weight coefficient” may be changed according to the number of times of recognition or the number of times of viewing, time, or the like.
A similar effect can be obtained even if the user-adaptive recognition dictionary is created using either the viewing history word or the past recognition word.

If hard resources are scarce, instead of weighting, for example, a new recognition word of a genre in which a program less frequently used by the user is classified may be deleted.
Further, when the user searches for a program to be broadcasted in the future, it is assumed that the program name does not utter, and for example, as shown in FIG. 13, a word whose “type” is the program name is not registered. It does not matter.

  Also, as the weighted user-adaptive recognition dictionary, it is possible to create a plurality of dictionaries having different weights according to the types of words. Further, by inputting a user recognition signal for recognizing a user to the dictionary creating unit 103, a recognition dictionary may be created for each user, and a plurality of user adaptive recognition dictionaries may be created.

  Here, for example, when creating a user-adaptive recognition dictionary in which the word “performer” can be easily recognized, the priority in voice recognition is set higher than the words “genre” and “program name”. , Can be weighted.

Also, with regard to the word “performer”, it is also possible to determine the priority in speech recognition by referring to the viewing history word and the frequency of use of the past recognition word and adapting it to the user's preference.
In addition, among the viewing history words and the past recognition words, words that are already included in the basic recognition dictionary are likely to be used again by the user, so that the weight for increasing the priority in voice recognition is high. By setting the coefficients, it is possible to create a weighted user adaptive recognition dictionary suitable for the usage situation of the user.

  In the present embodiment, the dictionary creation unit 103 creates three recognition dictionaries: a basic recognition dictionary, a user-adaptive recognition dictionary, and a weighted user-adaptive recognition dictionary, but is not limited thereto. Not something. For example, two recognition dictionaries, a basic recognition dictionary and any one of a user adaptive recognition dictionary and a weighted user adaptive recognition dictionary, may be created. Alternatively, a plurality of user adaptive recognition dictionaries and a plurality of weighted user adaptive recognition dictionaries may be created, and four or more recognition dictionaries may be created.

  Further, the operation of the offline processing step of creating a dictionary is executed at predetermined intervals, for example, once a day, but the execution time may be determined based on the viewing history.

  Next, the operation of the online processing step will be described. FIG. 10 is a flowchart showing the operation of the online processing step in the program guide device.

  The dictionary selecting unit 108 displays a plurality of recognition dictionaries stored in the dictionary storage unit 107 on, for example, a screen as shown in FIG. 11 and receives selection of a recognition dictionary by a user (step S201).

  Next, the dictionary selection unit 108 outputs the recognition dictionary selected by the user to the voice recognition unit 109 (Step S202). Note that the basic recognition dictionary can be output to the voice recognition unit 109 as an initial setting.

  Here, by displaying the type of the recognition dictionary selected by the user on a part of the television screen, it is possible to encourage the user to input voice contents in accordance with the contents of the recognition dictionary.

  Next, the voice recognition unit 109 recognizes a specific word from the input voice with reference to the recognition dictionary selected by the user (step S203). Here, a description will be given of a recognition result when the user utters "I want to see sports" while the user selects the weighted user adaptive recognition dictionary as shown in FIG. FIG. 12A is a diagram showing a part extracted from the weighted user adaptive recognition dictionary shown in FIG. 7, and FIG. 12B is a diagram showing an example of a recognition result. The speech recognition unit 109 first calculates a score as in general speech recognition. Here, as shown in FIG. 12 (b), it is assumed that the scores of “sports paradise” are 500, “sports” is 300, and “whole sports” are 200 as shown in FIG. The speech recognition unit 109 calculates the final score by multiplying the calculated score by the “target weight coefficient” and the “history weight coefficient” registered in the weighted user adaptive recognition dictionary. As a result, the voice recognition unit 109 recognizes “sports” as a final recognition result based on the final score.

  Next, the control unit 112 searches for a program by referring to the word recognized by the voice recognition unit 109 (hereinafter referred to as a recognized word) and the EPG data stored in the EPG storage unit 110 (step S204). . That is, the control unit 112 searches the EPG data stored in the EPG storage unit 110 based on the recognition word. Then, the control unit 112 determines whether or not the recognized word exists in the EPG data (Step S205).

  As a result, if the recognized word is included in the EPG data (YES in step S205), control unit 112 guides the corresponding program (step S206).

  On the other hand, when the recognized word is not included in the EPG data (NO in step S205), the control unit 112 determines that the program in which the talent C appears is “current and one week Are not broadcasted within "or" Movie K is not currently being broadcast "(step S207).

  Here, since the recognition dictionary is created using the viewing history and the recognition history of the user, if the words recognized by speech are included in the recognition dictionary but are not included in the EPG data, an error in speech recognition Otherwise, the control unit 112 determines that it is possible to reply to the presence or absence of the corresponding program.

  According to the above configuration, a word that is not included in the latest EPG data can be treated as a recognition target word by creating a recognition dictionary with reference to the EPG data, the user's viewing history, and the voice recognition history. When a program search is performed, an error in voice recognition and the presence or absence of search data can be identified, and a user-friendly voice interface can be realized.

  In addition, since the recognition dictionary is created by weighting the registered words and adding “target weight coefficient”, “history weight coefficient”, and the like, the recognition rate of speech recognition can be improved.

  Furthermore, since a plurality of recognition dictionaries are created and the user is allowed to select a recognition dictionary to be used, a user who thinks that the recognition rate of the voice recognition unit 109 is low is considered to be optimal from the plurality of dictionaries. By selecting a dictionary, the recognition rate can be improved and an easy-to-use interface can be realized.

(Embodiment 2)
In the second embodiment, in addition to the recognition dictionary for the user to search for programs to be broadcasted in the future described in the first embodiment, programs previously recorded (or watched) A case will be described in which a recognition dictionary for performing a search is created and the search is performed according to the purpose.

  FIG. 14 is a block diagram showing a configuration of a program guide device including a speech recognition dictionary creating device according to Embodiment 2 of the present invention. The same parts as those in the first embodiment are denoted by the same reference numerals, and detailed description will be omitted.

This program guide device includes a program / EPG receiving unit 201, a program / EPG storing unit 202, and a recording information storing unit 203, instead of EPG receiving unit 101 and EPG storing unit 110 of the first embodiment as shown in FIG. It has. Also, the operations of the dictionary creation unit 204, the dictionary selection unit 205, and the control unit 206 are different from those of the first embodiment.
The program / EPG receiving unit 201 separates EPG data and program data from the input television signal. The program / EPG storage unit 202 is a storage unit that stores EPG data and program data. The recording information storage unit 203 is a storage unit that stores information on a recorded program. Here, for example, as shown in FIG. 15, “notation” indicating a recorded program, “viewing date and time” indicating the date and time when the program was recorded, “type” indicating the genre of the program, and recording of the program “Reservation / accumulation” indicating whether the user has made the reservation or the system has made the reservation is recorded. Note that the system makes a recording reservation when, for example, all the programs including the keyword specified by the user are recorded.

The control unit 206 outputs information on the program to be recorded to the recording information storage unit 203.
The dictionary creation unit 204 uses the “target weight coefficient” of the weighted user-adaptive recognition dictionary described in Embodiment 1 as a recognition dictionary for performing a search on previously recorded (or watched) programs. For example, a weighted user-adaptive recognition dictionary as shown in FIG. 16 that is changed based on the utterance frequency survey result or the like shown in FIG. 6 is created. Here, as the “target weight coefficient”, “1.11” is used for a word whose “type” is a program name (unviewed program), and “1” is used for a word whose “type” is a program name (viewed program). .30 "," 1.30 "for words whose" type "is date and time," 1.03 "for words whose" type "is broadcasting station / channel, and" 1.03 "for words whose" type "is genre. 1.49 "and" 1.07 "are given to words whose" type "is a performer. At this time, the dictionary creation unit 204 determines whether the program name is an unviewed program or a watched program based on the “reservation / storage” information stored in the recording information storage unit 203. That is, when the system has made a recording reservation for a program, it determines that the program has not been viewed, and determines that the program has been already viewed by the user.

  The dictionary selection unit 205 receives an instruction from the user whether the search target is a program to be broadcasted in the future or a previously recorded (or watched) program, and selects a recognition dictionary according to the search target. . That is, when the search target is a program to be broadcasted in the future, the weighted user-adaptive recognition dictionary with a large genre weight described in Embodiment 1 shown in FIG. In the case of a program that has been watched or recorded in the above-mentioned manner, for example, a weighted user-adaptive recognition dictionary with a large weight of the program name as shown in FIG. 16 is selected.

Next, the operation of the online processing step of the program guide device configured as described above will be described. FIG. 17 is a flowchart showing the operation of the online processing step in the program guide device. Note that the operation of the off-line processing step is the same as that of the first embodiment, and a description thereof will be omitted.
The dictionary selection unit 205 receives an instruction from the user as to whether the search target is a program to be broadcasted in the future or a program that has been viewed or recorded before (step S301). If the search target is a program to be broadcasted in the future, a weighted user-adaptive recognition dictionary having a large genre weight is selected as a recognition dictionary used by the voice recognition unit 109 (step S302). On the other hand, if the search target is a previously watched or recorded program, a weighted user-adaptive recognition dictionary having a large program name weight is selected as a recognition dictionary used by the voice recognition unit 109 (step S303).

Subsequent operations (steps S203 to S207) are the same as in the first embodiment.
As described above, a plurality of recognition dictionaries are created depending on whether the search target is a program to be broadcasted in the future or a program recorded (or watched) before, and the recognition dictionary used in the voice recognition unit 109 is generated. Since the dictionary is selected, the recognition rate can be improved, and an easy-to-use interface can be realized.

  In the present embodiment, “reservation / accumulation” indicating whether the user has made a recording reservation for the program or the system has made recording is recorded as recording information stored in the recording information storage unit 203. It is not limited to this. For example, information indicating whether a program has been reserved for recording or whether the user has directly performed a recording operation, information indicating whether the user has actually viewed or not, and recording of the program The information may indicate whether the operation has been performed by using the EPG or by specifying the date and time. Further, a plurality of pieces of such information may be used.

(Embodiment 3)
In the first embodiment, when the recognition dictionary is updated, the newly created recognition dictionary and the old recognition dictionary are exchanged. In the third embodiment, the recognition dictionary is partially updated and registered in the recognition dictionary. A case will be described in which words registered before a predetermined period are deleted from among words that have been registered.

  FIG. 18 is a block diagram showing a configuration of a program guide device including a speech recognition dictionary creation device according to Embodiment 3 of the present invention. The same parts as those in the first embodiment are denoted by the same reference numerals, and detailed description will be omitted.

  This program guide device includes a word deleting unit 301 in addition to the configuration of the first embodiment as shown in FIG. Also, the operations of the dictionary creation unit 303 and the control unit 302 are different from those of the first embodiment.

The dictionary creation unit 303 creates a basic recognition dictionary, a user-adaptive recognition dictionary, and a weighted user-adaptive recognition dictionary as in the first embodiment, and then stores the basic recognition dictionary stored in the dictionary storage unit 107. A part different from the dictionary, the user adaptive recognition dictionary, and the weighted user adaptive recognition dictionary is partially updated.
The word deletion unit 503 refers to the date and time of the clock 111, the program content output from the control unit 112, and the past recognized words output from the recognized word storage unit 105, and selects one of the words registered in the recognition dictionary. , Delete words registered before a predetermined period (for example, one week).

Next, the operation of the offline processing step of the program guide device configured as described above will be described. FIG. 19 is a flowchart showing the operation of the online processing step in the program guide device. The operation in the online processing step is the same as in the first embodiment, and a description thereof will not be repeated.
The EPG receiving unit 101 to which the television signal has been input extracts the EPG data from the television signal and outputs it to the word extraction unit 102 and the EPG storage unit 110 (step S401).

  Next, the word extracting unit 102 extracts “program name”, “performer”, and “genre” from the EPG data as new recognition words to be keywords for voice recognition (step S402).

  Next, the dictionary creating unit 103 generates a new recognition word input from the word extraction unit 102, a fixed word stored in the fixed word storage unit 104, and a reading kana necessary for speech recognition stored in the fixed word storage unit 104. A basic recognition dictionary is created in the same manner as a conventional recognition dictionary using information such as part of speech (step S403).

  Next, the dictionary creation unit 103 refers to the viewing history word input from the viewing history word storage unit 106 and the past recognition word input from the recognition word storage unit 105, and determines whether a word other than the word registered in the basic recognition dictionary is used. If it exists, the word is added to the word registered in the basic recognition dictionary to create a user adaptive recognition dictionary as shown in FIG. 5, for example. Further, the dictionary creating unit 103 determines the weight according to the type of the word, and determines whether the word is stored as a past recognized word in the recognized word storage unit 105 and whether it is stored as a viewing history word in the viewing history word storage unit 106. A weighted user-adaptive recognition dictionary that stores, for example, as shown in FIG. 7, “target weight coefficient” and “history weight coefficient” in addition to “notation”, “read”, and “type” as shown in FIG. Is created (step S404).

  Next, the dictionary creation unit 103 compares the recognition dictionaries (basic recognition dictionary, user adaptive recognition dictionary, and weighted user adaptive recognition dictionary) stored in the dictionary storage unit 502 with the newly created recognition dictionary. Is obtained, and words that are not included in the already stored recognition dictionaries are registered in the dictionary storage unit 107 from the newly created recognition dictionaries, thereby partially replacing and updating the recognition dictionaries (step S405). ).

  Next, the word deletion unit 301 uses the date and time of the clock 111 and the date and time attached to the word in the recognition dictionary to delete the updated word from the recognition dictionary a predetermined period (for example, one week) before. . However, the period for deletion can be set by the user. In addition, words included in the user-adaptive recognition dictionary can be deleted for a longer period than the basic recognition dictionary. Further, using the user's viewing history, words related to programs viewed in the past can be deleted. It can also be excluded from deletion. Furthermore, it is also possible to change the period of deletion according to the type or genre of the word.

  When the hard resources are scarce, instead of weighting, for example, a new recognition word of a genre in which a program less frequently used by the user is classified may be deleted.

  Also, as the weighted user-adaptive recognition dictionary, it is possible to create a plurality of dictionaries having different weights according to the types of words. Alternatively, a viewing history may be created for each user, and a plurality of user-adaptive recognition dictionaries may be created.

A similar effect can be obtained even if the user-adaptive recognition dictionary is created using either the viewing history word or the past recognition word.
The updated date and time is attached to the updated word. If the same word exists in the old recognition dictionary, the date and time at the time of updating is attached again.

  According to the above configuration, a new recognition dictionary is created with reference to the EPG data, the user's viewing history, and the voice recognition history, and the previously created recognition dictionary is partially replaced to be included in the latest EPG data. Words that are not recognized can be treated as words to be recognized, and when a program search is performed, errors in voice recognition and the presence or absence of search data can be identified, and a user-friendly voice interface can be realized. .

  Furthermore, since a plurality of recognition dictionaries are created and the user is allowed to select a recognition dictionary to be used, a user who thinks that the recognition rate of the voice recognition unit 109 is low is considered to be optimal from the plurality of dictionaries. By selecting a dictionary, the recognition rate can be improved and an easy-to-use interface can be realized.

(Embodiment 4)
In the first to third embodiments, the case where a television broadcast program is searched has been described. In the fourth embodiment, a case where music content is searched will be described.

FIG. 20 is a block diagram showing a configuration of a music guidance device including a speech recognition dictionary creation device according to Embodiment 4 of the present invention. The same parts as those in the first embodiment are denoted by the same reference numerals, and detailed description will be omitted.
As shown in FIG. 20, the music guide device includes a music data / music information / viewing history storage unit 401 instead of the EPG receiving unit 101, the viewing history word storage unit 106, and the EPG storage unit 110 of the first embodiment. ing. The operations of the word extraction unit 402, the dictionary creation unit 403, and the control unit 404 are different from those of the first embodiment.

  The music data / music information / viewing history storage unit 401 is a storage unit that stores music data, music information, and viewing history. Here, as the music information, for example, as shown in FIG. 21, "music title" indicating a music title, "artist" indicating a player of the music, "album" indicating an album name in which the music is recorded, A “genre” indicating the genre of the music, and “details” indicating the playing time and the release date of the music are recorded.

Further, as the viewing history, for example, as shown in FIG. 22, the “music title” of the music, the “viewing period” indicating the period during which the music was viewed, the “viewing frequency” indicating the number of times the music was viewed, and the like. A “storage group” indicating where music was stored from is recorded.
The word extraction unit 402 extracts words for new recognition, which are words to be recognized, such as “music title”, “artist”, “album”, and “genre”, from the music information.

  The dictionary creation unit 403 adds information such as a reading kana and a part of speech necessary for speech recognition to the new recognition word extracted by the word extraction unit 402 and the fixed words stored in the fixed word storage unit 104. A basic recognition dictionary is created in the same manner as in the first embodiment. Further, the dictionary creating unit 403 refers to the past recognized words stored in the recognized word storage unit 105, and if a word other than the word registered in the basic recognition dictionary exists, the word registered in the basic recognition dictionary To create a user-adaptive recognition dictionary.

  Further, the dictionary creating unit 403 performs weighting according to whether the word is of a type such as “music title”, “artist”, “album”, “genre”, and determines a “target weight coefficient”. At this time, the dictionary creating unit 403 determines the “target weight coefficient” by making different weights for each type according to whether the music has not been viewed or has been viewed.

  Further, the dictionary creation unit 403 determines whether the word is stored as a previously recognized word in the recognized word storage unit 105, whether the word is related to music having a viewing history, and whether the user has stored the music. Weighting is performed according to whether the system is a system, and a “history weighting coefficient” is determined. Then, the dictionary creating unit 403 creates a weighted user-adaptive recognition dictionary that stores “target weight coefficient” and “history weight coefficient” in addition to “notation”, “read”, and “type”.

  Here, if the music has not been watched yet, the word “type” is a music title, as shown in FIG. 23, and “1.1”, “type” is an artist. "1.2" is assigned to the word, "1.1" is assigned to the word whose "type" is the album, and "1.3" is assigned to the word whose "type" is the genre. On the other hand, if the music has already been viewed, the word “type” is “1.3” for the word whose title is the music title, and “1.3” for the word whose type is the artist as the “target weight coefficient”. , "1.2" is given to words whose "type" is an album, and "1.3" is given to words whose "type" is a genre.

In addition, as a “history weighting coefficient”, “1.2” is stored for words stored as past recognized words in the recognized word storage unit 105, and “1.1” is stored for words related to music having a viewing history. "1.2" is given to words relating to music that is a user. At this time, if there is more than one, a coefficient is added to make a “history weighting coefficient”.
The operations of the online processing step of creating a dictionary and the offline processing step of recognizing input speech using the created dictionary and searching for music in the music guide device configured as described above are described in the embodiment. 1 and the description is omitted.

According to the above configuration, a word that is not included in the latest music information can be treated as a recognition target word by creating a recognition dictionary with reference to music information, a user's viewing history, and a voice recognition history. When a program search is performed, an error in voice recognition and the presence or absence of search data can be identified, and a user-friendly voice interface can be realized.
In addition, since the recognition dictionary is created by weighting the registered words and adding “target weight coefficient”, “history weight coefficient”, and the like, the recognition rate of speech recognition can be improved.

  Furthermore, since a plurality of recognition dictionaries are created and the user is allowed to select a recognition dictionary to be used, a user who thinks that the recognition rate of the voice recognition unit 109 is low is considered to be optimal from the plurality of dictionaries. By selecting a dictionary, the recognition rate can be improved and an easy-to-use interface can be realized.

  In the present embodiment, for the “history weighting coefficient”, whether the word is stored in the recognized word storage unit 105 as a past recognized word, whether the word is related to music having a viewing history, or not is stored. Although weighting is performed according to whether the user is a user or a system, weighting may be performed using only one of them. Further, a “weight coefficient” may be individually added to the dictionary. Further, the “weight coefficient” may be changed according to the number of times of recognition or the number of times of viewing, time, or the like.

  Further, in the present embodiment, it is assumed that the search is performed without dividing the search target into unviewed or already viewed, but the present invention is not limited to this. For example, when performing a search by dividing the search target into unviewed or already watched, different "target weights" are used when searching for the search target separately and when searching without separating the search target. A coefficient may be used.

(Embodiment 5)
In the fourth embodiment, a case in which music content is searched has been described. In the fifth embodiment, a case in which store (restaurant) data is searched will be described.

  FIG. 24 is a block diagram showing a configuration of an information search device including a speech recognition dictionary creation device according to Embodiment 5 of the present invention. The same parts as those in the first embodiment are denoted by the same reference numerals, and detailed description will be omitted.

As shown in FIG. 24, this information search device includes a restaurant data storage unit 501 and a usage information storage unit 504 instead of the EPG reception unit 101, the viewing history word storage unit 106, and the EPG storage unit 110 of the first embodiment. Have. Also, the operations of the word extraction unit 502, the dictionary creation unit 503, and the control unit 505 are different from those of the first embodiment.
The restaurant data storage unit 501 is a storage unit that stores restaurant data. Here, for example, as shown in FIG. 25, “store name” indicating the name of the restaurant, “genre 1” indicating the general genre of the restaurant, “genre 2” indicating the detailed genre of the restaurant, and the restaurant "Main menu" indicating the menu handled by the "" and "business hours" indicating the business hours of the restaurant are recorded.

The usage information storage unit 504 is a storage unit that stores usage information of a restaurant used by the user. Here, for example, as shown in FIG. 26, the “store name” of the restaurant, the “use period” indicating the period during which the user has used the restaurant, and the “use count” indicating the number of times the user has used the restaurant , And a “use time zone” indicating the time zone in which the restaurant was used by the user.
The word extraction unit 502 extracts new recognition words that are words to be recognized, such as “store name”, “genre 1”, “genre 2”, and “main menu”, from the restaurant data.

  The dictionary creation unit 503 adds the information such as the reading kana and the part of speech necessary for speech recognition to the new recognition word extracted by the word extraction unit 502 and the fixed words stored in the fixed word storage unit 104. A basic recognition dictionary is created in the same manner as in the first embodiment. Further, the dictionary creating unit 503 refers to the past recognized words stored in the recognized word storage unit 105 and, if there is a word other than the word registered in the basic recognition dictionary, adds the word to the word registered in the basic recognition dictionary. To create a user-adaptive recognition dictionary.

Further, the dictionary creation unit 503 performs weighting according to the type of the word such as “store name”, “genre 1”, “genre 2”, “main menu”, and the like, and sets “object weight coefficient”. decide. At this time, the dictionary creating unit 503 determines the “target weight coefficient” by making different weights for each type according to whether or not the user has used the restaurant.
Further, the dictionary creating unit 503 performs weighting according to whether the word is stored in the recognized word storage unit 105 as a past recognized word, and determines a “history weight coefficient”. Then, the dictionary creating unit 503 creates a weighted user-adaptive recognition dictionary that stores “target weight coefficient” and “history weight coefficient” in addition to “notation”, “read”, and “type”.

  Here, when the restaurant is not used by the user, as the “target weight coefficient”, as shown in FIG. A word of genre 1 is given “1.4”, a word of “type” is genre 2 is given “1.3”, and a word of “type” is a main menu is given “1.3”. I have. On the other hand, when the restaurant is used by a user, the word “type” is “1.4” as a “target weight coefficient”, and the word “type” is genre 1 as a word. "1.4" is given to words whose "1.3" and "type" are genre 2, and "1.2" is given to words whose main menu is "type".

In addition, “1.5” is given as a “history weighting coefficient” to a word stored in the recognized word storage unit 105 as a past recognized word.
The operation of the online processing step of creating a dictionary in the information search device configured as described above, and the operation of the offline processing step of recognizing input speech using the created dictionary and searching for restaurants are described in the following. The description is omitted because it is similar to the first embodiment.

According to the above configuration, a word that is not included in the latest restaurant data can be treated as a recognition target word by creating a recognition dictionary with reference to the user's usage information and the voice recognition history. When a store search is performed, an error in voice recognition and the presence or absence of search data can be identified, and a user-friendly voice interface can be realized.
In addition, since the recognition dictionary is created by weighting the registered words and adding “target weight coefficient”, “history weight coefficient”, and the like, the recognition rate of speech recognition can be improved.

Furthermore, since a plurality of recognition dictionaries are created and the user is allowed to select a recognition dictionary to be used, a user who thinks that the recognition rate of the voice recognition unit 109 is low is considered to be optimal from the plurality of dictionaries. By selecting a dictionary, the recognition rate can be improved and an easy-to-use interface can be realized.
In the present embodiment, the “history weighting coefficient” is weighted according to whether the word is stored in the recognized word storage unit 105 as a past recognized word. It may be changed by a band or the like.

  Further, in the present embodiment, it is assumed that a search is performed without dividing a search target into those who have or have not used the search object. However, the present invention is not limited to this. For example, when searching for a search target with or without experience in using it, a different "target weight coefficient" is used for searching for the search target separately and for searching without separating the search target. No problem.

  As described above, the speech recognition dictionary creation device and the information search device according to the present invention can create a recognition dictionary with an improved recognition rate of speech recognition. For example, a program guide device, a music guide device, and a car navigation device It is useful for an information search device such as a system to search for information by a voice signal.

FIG. 1 is a block diagram illustrating a configuration of a program guide device including a speech recognition dictionary creation device according to Embodiment 1 of the present invention. FIG. 6 is a diagram illustrating an example of data stored in a recognition word storage unit. FIG. 6 is a diagram illustrating an example of data stored in a viewing history word storage unit. It is a figure showing the example of data of a basic recognition dictionary. It is a figure showing the example of data of a user adaptive type recognition dictionary. It is a figure showing the result of having investigated the frequency with which the kind of word was uttered in each utterance condition. It is a figure showing an example of data of a weighted user adaptive type recognition dictionary. It is a flowchart which shows operation | movement of the offline processing process in a program guide apparatus. FIG. 4 is a diagram illustrating an example of EPG data. It is a flowchart which shows operation | movement of the online processing process in a program guide apparatus. FIG. 9 is a diagram illustrating an example of a screen for selecting one recognition dictionary from a plurality of recognition dictionaries. (A) is a figure which extracted a part from the weighted user adaptive type recognition dictionary shown in FIG. 7, and (b) is a figure which shows the example of a recognition result. It is a figure which shows the example of data of the user adaptive recognition dictionary in which the word whose "type" is a program name is not registered. FIG. 9 is a block diagram showing a configuration of a program guide device including a speech recognition dictionary creation device according to Embodiment 2 of the present invention. FIG. 4 is a diagram illustrating an example of data stored in a recording information storage unit. FIG. 13 is a diagram illustrating an example of data of a weighted user adaptive recognition dictionary according to the second embodiment. 9 is a flowchart illustrating an operation of an online processing step in the program guide device according to the second embodiment. FIG. 9 is a block diagram showing a configuration of a program guide device including a speech recognition dictionary creation device according to Embodiment 3 of the present invention. 13 is a flowchart illustrating an operation of an online processing step in the program guide device according to the third embodiment. FIG. 13 is a block diagram showing a configuration of a music guidance device including a speech recognition dictionary creation device according to Embodiment 4 of the present invention. It is a figure showing the example of data of music information stored in music data / music information / viewing history accumulation part. FIG. 4 is a diagram illustrating an example of data of a viewing history stored in a music data / music information / viewing history storage unit. 129 is a diagram illustrating an example of “target weight coefficients” for each “kind” in Embodiment 4. FIG. FIG. 14 is a block diagram illustrating a configuration of an information search device including a speech recognition dictionary creation device according to Embodiment 5 of the present invention. It is a figure showing an example of data stored in a restaurant data storage part. FIG. 4 is a diagram illustrating an example of data stored in a usage information storage unit. 157 is a diagram illustrating an example of “target weight coefficients” for each “kind” in Embodiment 5. FIG. FIG. 11 is a block diagram illustrating a configuration of a program guide device including a conventional voice recognition dictionary creation device.

Explanation of reference numerals

101 EPG receiving unit 102, 402, 502 Word extraction unit 103, 204, 303, 403, 503 Dictionary creation unit 104 Fixed word storage unit 105 Recognized word storage unit 106 Viewing history word storage unit 107 Dictionary storage unit 108, 205 Dictionary selection unit 109 Voice recognition unit 110 EPG storage unit 111 Clock 112, 206, 302, 404, 505 Control unit 201 Program / EPG reception unit 202 Program / EPG storage unit 203 Recording information storage unit 301 Word deletion unit 401 Music data / music information / viewing History storage unit 501 Restaurant data storage unit 504 Usage information storage unit

Claims (21)

A speech recognition dictionary creating apparatus for creating a recognition dictionary for performing speech recognition on data related to program information or music content ,
Viewing history word acquisition means for acquiring a viewing history word that is a word to be recognized based on the viewing experience of the data of the user,
Viewing history word accumulation means for accumulating the acquired viewing history words,
Fixed word storage means for storing a predetermined fixed word set in advance;
Word extraction means for extracting a new recognition word, which is a word to be recognized, from input program information or data relating to music content;
Dictionary creation means for creating a plurality of recognition dictionaries having different contents from the new recognition word, the fixed word, and the viewing history word ,
A speech recognition dictionary creating apparatus , comprising: dictionary storage means for storing the created recognition dictionary. The speech recognition dictionary creation device according to claim 1, wherein the dictionary creation unit creates a plurality of the recognition dictionaries having different contents based on an attribute of a word registered in the recognition dictionary. Before Symbol dictionary creating means, the word for the new recognition, the fixed word, and to determine a new registration word is a word to be registered in the recognition dictionary from the viewing history word, reading of the newly registered word to the newly registered word The speech recognition dictionary creation device according to claim 1, wherein the recognition dictionary is created by adding information indicating a kana and a part of speech. The dictionary creation means determines the new recognition word and the fixed word, and words not included in the new recognition word and the fixed word in the viewing history words as the newly registered word. 2. The speech recognition dictionary creating device according to claim 1, wherein: The voice recognition dictionary creation device further includes:
A recognized word acquiring means for acquiring a past recognized word that is a word recognized by voice recognition;
Comprising a past recognition word storage means for storing the past recognition words,
The dictionary creation means, said word for the new recognition, said fixed word, the viewing history word, and claims 1, wherein said recognition dictionary to a plurality create having different contents from the past recognized word 5. The speech recognition dictionary creating device according to any one of 4 . The dictionary creating means determines a newly registered word to be registered in a recognition dictionary from the new recognition word, the fixed word, the viewing history word, and the past recognized word, and determines the newly registered word in the newly registered word. The speech recognition dictionary creation device according to claim 5, wherein the recognition dictionary is created by adding information indicating a word reading kana and a part of speech. The dictionary creation unit may include the new recognition word and the fixed word, and the viewing history word and the past recognition word, and the words not included in the new recognition word and the fixed word, are added to the newly registered word. The speech recognition dictionary creation device according to claim 6, wherein: The dictionary creation means weights the newly registered word based on whether or not the word is included in the viewing history word, and adds the information on the weighting to create the recognition dictionary. The speech recognition dictionary creation device according to claim 1 or 5, wherein The dictionary creation unit weights the newly registered word based on whether or not the word is included in the past recognition word, and adds the information on the weight to create the recognition dictionary. The speech recognition dictionary creation device according to claim 5, wherein The dictionary creation unit weights the newly registered word based on whether or not the word is included in the viewing history word and the past recognition word, and adds information regarding the weighting to the recognition dictionary. The speech recognition dictionary creating device according to claim 5, wherein the dictionary is created. The input data is structured data,
The word extracting means, together with the new recognition word, extracts an attribute of the new recognition word,
The dictionary creation means, wherein performs weighting of the newly registered word on the basis of the attributes of the word for the new recognition claim 1 or claim, characterized in that to create the recognition dictionary by adding information about the weighting 5. The speech recognition dictionary creation device according to 5. The speech recognition dictionary according to claim 11 , wherein the dictionary creation unit creates a plurality of the recognition dictionaries in which weights of the newly registered words are different based on attributes of the new recognition word, according to purposes. Creation device. An instruction receiving unit that receives an instruction regarding a use purpose of the recognition dictionary from the user,
The dictionary creation unit creates the recognition dictionary by weighting the newly registered word based on the attribute of the new recognition word in accordance with the instruction on the purpose of use received by the instruction reception unit. The speech recognition dictionary creating device according to claim 11, wherein The voice recognition dictionary creation device further includes:
Comprising user information acquisition means for acquiring user identification information for identifying the user,
The dictionary creation means, the user creating the recognition dictionary for each recognition information speech recognition dictionary creating apparatus according to claim 1 or claim 5 wherein. The voice recognition dictionary creation device further includes:
A user information acquisition unit for acquiring user identification information for identifying the user,
The speech recognition dictionary creation device according to claim 11 , wherein the dictionary creation unit creates the recognition dictionary by weighting the newly registered word for each of the user recognition information. The voice recognition dictionary creation device further includes:
EPG receiving means for separating EPG data as an electronic program guide from a television signal,
The action of the user is watching a program,
The viewing history word acquiring means acquires the viewing history word based on the user's viewing ,
The word extracting unit, according to claim 1 speech recognition dictionary creating apparatus, wherein the extracting the word for a new recognition from the EPG data for each item that is the attribute. 17. The speech recognition according to claim 16 , wherein the dictionary creating unit weights the newly registered word based on the item of the EPG data, and creates the recognition dictionary by adding information relating to the weighting. Dictionary creation device. The dictionary creating means includes a program recognition dictionary for future programs in which the weight for the registered word whose item is a genre is greater than the weight for the registered word whose item is a title, and the registered word whose item is a title. 18. The speech recognition dictionary creation device according to claim 17, wherein a speech recognition dictionary for a recorded program is created in which the weight of the recorded program is weighted more than the weight of the registered word whose item is the genre. An information search device for creating a recognition dictionary for performing voice recognition on program information or data related to music content, and searching for information by a voice signal generated by a user's speech,
Viewing history word acquisition means for acquiring a viewing history word that is a word to be recognized based on the viewing experience of the data of the user,
Viewing history word accumulation means for accumulating the acquired viewing history words,
Fixed word storage means for storing a predetermined fixed word set in advance;
Word extraction means for extracting a new recognition word, which is a word to be recognized, from input program information or data relating to music content;
Dictionary creation means for creating a plurality of recognition dictionaries having different contents from the new recognition word, the fixed word, and the viewing history word ,
Dictionary storage means for storing the created recognition dictionary;
Voice recognition means for recognizing a specific word from the voice signal by referring to the recognition dictionary;
Data storage means for storing input data;
A control unit that searches the data for the word recognized by the voice recognition unit and outputs corresponding information. An information search method for creating a recognition dictionary for performing voice recognition on program information or data on music content, and searching for information by a voice signal generated by a user's speech,
A viewing history word obtaining step of obtaining a viewing history word that is a word to be recognized based on the viewing experience regarding the data of the user,
A word extraction step of extracting a new recognition word, which is a word to be recognized, from input program information or data relating to music content;
A dictionary creation step of creating a plurality of the recognition dictionaries having different contents from the new recognition word, a predetermined fixed word set in advance, and the viewing history word ;
A voice recognition step of recognizing a specific word from the voice signal by referring to the recognition dictionary;
A search step of searching the input data for a word recognized in the voice recognition step , and outputting corresponding information. A program for creating a recognition dictionary for performing voice recognition on program information or data relating to music content and searching for information by a voice signal generated by a user's speech,
A viewing history word acquiring step of acquiring a viewing history word that is a word to be recognized based on the viewing experience of the user regarding the data,
A word extraction step of extracting a new recognition word, which is a word to be recognized, from input program information or data relating to music content;
A dictionary creation step of creating a plurality of the recognition dictionaries having different contents from the new recognition word, a predetermined fixed word set in advance, and the viewing history word ;
A voice recognition step of recognizing a specific word from the voice signal by referring to the recognition dictionary;
And a control step of searching the input data for the word recognized in the voice recognition step and outputting corresponding information.