JP2013196019A - Retrieval control device, output device, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To properly present useful information to a user.SOLUTION: A retrieval control device includes an extraction unit for extracting a word included in input first information, and a control unit that causes retrieval means to retrieve second information output as a retrieval result on the basis of retrieval conditions based on the word extracted by the extraction unit. The control unit has a first retrieval mode for widening retrieval conditions and a second retrieval mode for narrowing the retrieval conditions.

Description

  The present invention relates to a search control device, an output device, and a program.

  In recent years, a technique for retrieving information from a retrieval device such as a personal computer, a mobile phone, or a car navigation device has been known (see, for example, Patent Document 1).

JP 2007-017990 A

  However, in such a search device technology, for example, a search is performed based on a fixed search condition intentionally selected by the user, such as destination information. Since the search is based on a fixed search condition selected by the user, such a search apparatus may not be able to provide appropriate information to the user depending on, for example, the search condition.

  The present invention has been made to solve the above problems, and an object of the present invention is to provide a search control device, an output device, and a program that can appropriately provide useful information to the user.

  In order to solve the above problem, an embodiment of the present invention is based on an extraction unit that extracts a word included in input first information, and a search condition based on the word extracted by the extraction unit, A control unit that causes the search unit to search for second information output as a search result, wherein the control unit narrows the search condition and a first search mode in which the search condition is widened. And a second search mode.

  According to an embodiment of the present invention, an extraction unit that extracts a word included in the input first information, and second information that is output as a search result based on the word extracted by the extraction unit A control unit that causes the search means to search for, when the number of the search results based on the word is equal to or less than a predetermined number, the related word related to the word In accordance with the present invention, the search control device causes the search means to search for the second information.

  In addition, an embodiment of the present invention is an output device comprising: the search control device described above; and an output unit that outputs the second information searched by the search control device.

  One embodiment of the present invention is a program for causing a computer as a search control device to execute a program, in which the extraction unit extracts a word included in the input first information, and the control unit A control procedure for causing the search means to search for second information output as a search result based on a search condition based on the word extracted by the extraction procedure, and the control procedure includes the search A program including a first search mode for widening a condition and a second search mode for narrowing the search condition.

  According to the present invention, it is possible to appropriately provide useful information to the user.

It is a 1st figure which shows an example of a structure of the search device by this embodiment. It is a block diagram of the search device in a 1st embodiment. It is a flowchart which shows an example of a process of the topic provision mode in 1st Embodiment. It is a flowchart which shows an example of the process of conversation tracking mode in 1st Embodiment. It is a flowchart which shows the 1st modification of the topic provision mode in 1st Embodiment. It is a flowchart which shows the 2nd modification of the topic provision mode in 1st Embodiment. It is a block diagram of the search device in a 2nd embodiment. It is a flowchart which shows an example of the process of conversation tracking mode in 2nd Embodiment. It is a 2nd figure which shows an example of a structure of the search device by this embodiment. It is a block diagram of the search device in a 3rd embodiment. It is a flowchart which shows an example of the process of the conversation tracking mode in 3rd Embodiment. It is a flowchart which shows an example of the process of conversation tracking mode in 4th Embodiment. It is a block diagram of the search device in a 5th embodiment.

Hereinafter, a search control device and a search device (output device) according to an embodiment of the present invention will be described with reference to the drawings.
[First Embodiment]
FIG. 1 is a first diagram illustrating an example of a system configuration of a search device 1 according to the present embodiment.
In FIG. 1, a search device 1 is connected to a search server 4 via a network 3 (for example, the Internet). In addition, the search device 1 is connected to a sound collection microphone 2. Thereby, voice information (for example, voice signal of conversation) of the user (for example, two users (U 1, U 2)) is input from the sound collection microphone 2 to the search device 1.

The sound collecting microphone 2 picks up the voice of the user's conversation and converts it into a sound signal (voice information). The sound collection microphone 2 outputs a sound signal (voice information) obtained by converting the user's voice to the search device 1 as first information.
The search server 4 (search means) searches for information (second information) such as images, sounds, and sentences related to the keyword based on a keyword (search word) or a search expression based on the keyword. For example, the search server 4 may search for information (second information) as a search result via the network 3, or information as a search result from information stored in a database provided in the own device. You may search (2nd information).

FIG. 2 is a block diagram of the search device 1 according to the present embodiment.
In FIG. 2, the search device 1 (output device) includes a control unit 10 and a display unit 20. In the present embodiment, the control unit 10 functions as a search control device.
The control unit 10 controls each unit of the search device 1. The control unit 10 causes the search server 4 to search via the network 3 based on the voice information (first information) input from the sound collection microphone 2, and information such as an image (second information) as a search result. To get. Further, the control unit 10 causes the display unit 20 to display the search result acquired from the search server 4.
The control unit 10 includes a voice recognition unit 11 and a search control unit 12.

  The speech recognition unit 11 (extraction unit) extracts words included in the speech information (first information) input from the sound collection microphone 2. That is, the voice recognition unit 11 recognizes input voice information, converts it into text information (sentence), and extracts a word included in the text information. The voice recognition unit 11 supplies the extracted word to the search control unit 12.

The search control unit 12 (control unit) causes the search server 4 to search for information such as an image output as a search result based on the words extracted by the speech recognition unit 11. That is, the search control unit 12 causes the search server 4 to search for the second information based on the search condition based on the word extracted by the voice recognition unit 11. In addition, the search control unit 12 generates a search expression based on keywords (for example, nouns, proper nouns, etc.) based on the words extracted by the speech recognition unit 11, and information such as images based on the generated search expressions. To the search server 4. For example, the search control unit 12 generates a search expression in which keywords of the words included in the first information are sequentially added as a search condition, and based on the generated search expression, information such as an image is searched for by the search server 4. To search. That is, the search control unit 12 supplies the generated search expression to the search server 4 via the network 3 and acquires information such as images as search results from the search server 4 via the network 3. Note that the above-described keyword (search word) may be a character string (for example, an adjective + noun) or a sentence combining words.
Here, the search control unit 12 has the following two search modes.

  The topic provision mode (first search mode) is a search mode in which search conditions are expanded (expanded), and is a search mode having a wider search condition than a conversation tracking mode described later. Topic offer mode, for example, information that provides a topic in the user's conversation information such as images searched for the content that the user is currently interested in, using the words included in the conversation by the user such as a conference as a keyword As a search mode. This topic providing mode is an effective search mode for brainstorming in a conference. In the topic providing mode, the search control unit 12 combines keywords included in the newly input voice information (first information) with a logical sum (OR: OR) and adds them to the search formula.

  For example, in the topic providing mode, the search control unit 12 generates a search formula as shown in the following formula (1).

Search formula = KEY1 or KEY2 (1)
Here, KEY1 and KEY2 are symbols indicating keywords.

  When the speech recognition unit 11 newly extracts a keyword (KEY3) in the state of the search formula of the formula (1), the search control unit 12 searches the following formula (2) with the keyword (KEY3) added. Generate an expression.

  Search formula = KEY1 or KEY2 or KEY3 (2)

  The conversation tracking mode (second search mode) is a search mode in which search conditions are narrowed (reduced), and is a search mode with narrower search conditions than the topic provision mode. The conversation tracking mode is a search mode in which information such as images searched by narrowing down the content that the user wants to know is used as a keyword for words included in conversations by users such as meetings, for example. This conversation tracking mode is a search mode that is effective in the work of gathering opinions and information in a meeting. In the conversation tracking mode, the search control unit 12 combines keywords included in the newly input voice information (first information) by logical product (AND) and adds the keywords to the search formula.

  For example, in the conversation tracking mode, the search control unit 12 generates a search formula as shown in the following formula (3).

Search formula = KEY1 and KEY2 (3)
Here, KEY1 and KEY2 are symbols indicating keywords.

  When the speech recognition unit 11 newly extracts a keyword (KEY3) in the state of the search formula of this formula (3), the search control unit 12 searches the following formula (4) with the keyword (KEY3) added. Generate an expression.

  Search formula = KEY1 and KEY2 and KEY3 (4)

In addition, when the word extracted by the speech recognition unit 11 includes a predetermined specified word (predetermined word), the search control unit 12 determines the topic providing mode and the conversation tracking mode according to the specified word. And switch. For example, when the search control unit 12 is interested in information such as an image (image group) displayed on the display unit 20, the search control unit 12 switches from the topic providing mode to the conversation tracking mode. That is, when the word extracted by the speech recognition unit 11 includes a pronoun such as “what” or “this”, the search control unit 12 makes a transition from the topic providing mode to the conversation tracking mode.
In addition, for example, when the user is satisfied with the search result, the search control unit 12 shifts from the conversation tracking mode to the topic providing mode. That is, the search control unit 12 includes specified words that are found to be satisfied in the words extracted by the speech recognition unit 11, for example, nouns or pronouns such as “OK”, “around”, “it”, and the like. In this case, the conversation tracking mode is changed to the topic providing mode.

  In the topic providing mode, the search control unit 12 may acquire a lot of unnecessary search results because there is too much information such as images as search results. Therefore, in the present embodiment, the search control unit 12 excludes the keyword from the search expression when the number of search results exceeds a predetermined first predetermined number in the topic providing mode, and the search server Let 4 search again. In this case, for example, the search control unit 12 excludes keywords from the search formula in the oldest order (the oldest input order) extracted by the speech recognition unit 11. Here, excluding the above keywords from the search expression means to delete (exclude) some keywords from the keywords included in the search expression.

  In addition, the search control unit 12 may not be able to obtain a required search result in the conversation tracking mode. Therefore, in the present embodiment, the search control unit 12 excludes the keyword from the search formula when the number of search results is equal to or less than a predetermined second specified number in the conversation tracking mode, and the search server Let 4 search again. In this case, for example, the search control unit 12 excludes keywords from the search formula in the oldest order (the oldest input order) extracted by the speech recognition unit 11.

  As described above, the search control unit 12 selectively switches between the topic providing mode and the conversation tracking mode based on the words extracted by the voice recognition unit 11 and generates a search expression. The search control unit 12 supplies the generated search expression to the search server 4, and acquires information such as an image as a search result from the search server 4 via the network 3. The search control unit 12 supplies the search result acquired from the search server 4 to the display unit 20 and causes the display unit 20 to display the search result.

  The display unit 20 is a display device such as a liquid crystal display, for example. The display unit 20 displays information (second information) such as an image supplied from the search control unit 12.

Next, the operation of the search device 1 in this embodiment will be described.
FIG. 3 is a flowchart showing an example of processing in the topic provision mode in the present embodiment.
In this figure, first, the control unit 10 determines whether or not there is an audio input from the sound collection microphone 2 (step S101). That is, the control unit 10 determines whether voice information is input from the sound collection microphone 2 to the voice recognition unit 11. When it is determined that the voice information has been input (step S101: YES), the control unit 10 advances the process to step S102. When it is determined that the voice information is not input (step S101: NO), the control unit 10 returns the process to step S101 and repeats the process of step S101 until the voice information is input.

  Next, the control unit 10 performs voice recognition and morphological analysis (step S102). That is, the voice recognition unit 11 of the control unit 10 recognizes the input voice information, converts it into text information (sentence), and extracts words included in the text information. The voice recognition unit 11 supplies the extracted word to the search control unit 12.

  Next, the control unit 10 performs search word (keyword) addition and search processing (step S103). That is, the search control unit 12 of the control unit 10 combines the keywords included in the newly input voice information (first information) with a logical sum (OR) and adds the keywords to the search formula. The search control unit 12 supplies the generated search expression to the search server 4 via the network 3, and acquires information such as images as search results from the search server 4 via the network 3.

  Next, the control part 10 displays a search result on the display part 20 (step S104). That is, the search control unit 12 of the control unit 10 supplies information such as an image acquired as a search result to the display unit 20 for display. Accordingly, the display unit 20 displays information such as an image that is a search result supplied from the search control unit 12.

  Next, the control unit 10 determines whether or not a specified word is included in the input voice information (step S105). That is, the search control unit 12 of the control unit 10 determines whether or not a prescribed word (predetermined word) is included in the words extracted by the speech recognition unit 11. Here, as described above, the specified word is a pronoun such as “what” or “this”. When it is determined that the specified word (predetermined word) is included (step S105: YES), the search control unit 12 shifts the processing from the topic providing mode to the conversation tracking mode (step S111 in FIG. 4). If the search control unit 12 determines that the specified word (predetermined word) is included (step S105: NO), the search control unit 12 advances the process to step S106.

  Next, in step S106, the control unit 10 determines whether or not the search result is equal to or greater than a specified number (first specified number). That is, the search control unit 12 of the control unit 10 determines whether or not the number of search results (second information) acquired from the search server 4 is equal to or greater than a specified number (first specified number). If the search control unit 12 determines that the number of search results is greater than or equal to the specified number (first specified number) (step S106: YES), the search control unit 12 advances the process to step S107. In addition, when the search control unit 12 determines that the number of search results is less than the specified number (first specified number) (step S106: NO), the search control unit 12 returns the process to step S101.

  Next, in step S107, the control part 10 deletes the word contained in an old sentence (voice information), and returns a process to step S101. That is, the search control unit 12 of the control unit 10 excludes keywords from the search formula in the oldest order (the oldest input order) extracted by the speech recognition unit 11, and returns the process to step S101.

  As described above, in the topic providing mode in the present embodiment, the control unit 10 combines the keywords included in the newly input voice information (first information) with a logical sum (OR) and adds the keywords to the search formula. Then, the search server 4 is caused to search for information such as an image as a search result. Then, the control unit 10 causes the display unit 20 to display the search results, and excludes the keyword from the search formula when the number of search results is equal to or more than a predetermined first predetermined number, and the search server Let 4 search again.

FIG. 4 is a flowchart illustrating an example of processing in the conversation tracking mode in the present embodiment.
In this figure, first, the control unit 10 determines whether or not there is an audio input from the sound collection microphone 2 (step S111). That is, the control unit 10 determines whether voice information is input from the sound collection microphone 2 to the voice recognition unit 11. When it is determined that the voice information has been input (step S111: YES), the control unit 10 advances the process to step S112. When it is determined that the voice information is not input (step S111: NO), the control unit 10 returns the process to step S111 and repeats the process of step S111 until the voice information is input.

  Next, the control unit 10 performs voice recognition and morphological analysis (step S112). That is, the voice recognition unit 11 of the control unit 10 recognizes the input voice information, converts it into text information (sentence), and extracts words included in the text information. The voice recognition unit 11 supplies the extracted word to the search control unit 12.

  Next, the control unit 10 performs search word (keyword) addition and search processing (step S113). That is, the search control unit 12 of the control unit 10 combines the keywords included in the newly input voice information (first information) with the logical product (AND) and adds them to the search formula. The search control unit 12 supplies the generated search expression to the search server 4 via the network 3, and acquires information such as images as search results from the search server 4 via the network 3.

  Next, the control part 10 displays a search result on the display part 20 (step S114). That is, the search control unit 12 of the control unit 10 supplies information such as an image acquired as a search result to the display unit 20 for display. Accordingly, the display unit 20 displays information such as an image that is a search result supplied from the search control unit 12.

  Next, the control unit 10 determines whether or not a specified word is included in the input voice information (step S115). That is, the search control unit 12 of the control unit 10 determines whether or not a prescribed word (predetermined word) is included in the words extracted by the speech recognition unit 11. Here, the specified word is a noun or pronoun such as “OK”, “around”, “it”, etc., as described above. When the search control unit 12 determines that the specified word (predetermined word) is included (step S115: YES), the search control unit 12 shifts the process from the conversation tracking mode to the topic providing mode (step S101 in FIG. 3). If the search control unit 12 determines that the specified word (predetermined word) is included (step S115: NO), the search control unit 12 advances the process to step S116.

  Next, in step S116, the control unit 10 determines whether or not the search result is equal to or less than a specified number (second specified number). That is, the search control unit 12 of the control unit 10 determines whether or not the number of search results acquired from the search server 4 is equal to or less than a specified number (second specified number). If the search control unit 12 determines that the number of search results is equal to or less than the specified number (first specified number) (step S116: YES), the search control unit 12 advances the process to step S117. Moreover, the search control part 12 returns a process to step S111, when it determines with the number of search results exceeding a regulation number (2nd regulation number) (step S116: NO). In the present embodiment, the first specified number and the second specified number described above have a relationship of (first specified number> second specified number), and the first specified number and the second specified number The specified number is determined in advance so as to maintain an appropriate number of search results.

  Next, in step S117, the control part 10 deletes the word contained in an old sentence (audio | voice information), and returns a process to step S111. That is, the search control unit 12 of the control unit 10 excludes keywords from the search formula in the oldest order (the oldest input order) extracted by the speech recognition unit 11, and returns the process to step S111.

  As described above, in the conversation tracking mode according to the present embodiment, the control unit 10 combines the keywords included in the newly input voice information (first information) with the logical product (AND) and adds them to the search expression. Then, the search server 4 is caused to search for information such as an image as a search result. Then, the control unit 10 causes the display unit 20 to display the search results, and excludes the keyword from the search formula when the number of search results is equal to or less than a predetermined second predetermined number, and the search server Let 4 search again.

  As described above, the control unit 10 (search control device) in the present embodiment includes the voice recognition unit 11 and the search control unit 12. The voice recognition unit 11 extracts words included in the input voice information (first information). The search control unit 12 causes the search server 4 to search for information such as an image output as a search result based on the search condition based on the words extracted by the speech recognition unit 11. The search control unit 12 has a topic providing mode for widening the search conditions and a conversation tracking mode for narrowing the search conditions.

  Thereby, the search device 1 according to the present embodiment displays, on the display unit 20, for example, a search result obtained by searching for contents that the user is currently interested in from the conversation of the user in the topic providing mode. Then, in the conversation tracking mode, the search device 1 according to the present embodiment displays the searched search result on the display unit 20 so as to narrow down the content that the user wants to know. Therefore, the control unit 10 and the search device 1 according to the present embodiment can provide information narrowed down to contents that the user wants to know while providing a wide range of information to the user. For example, the control unit 10 and the search device 1 according to the present embodiment are not limited to fixed search conditions that are intentionally selected by the user, but are based on search conditions extracted from the user's natural conversations. It is possible to provide information that delves into the topics that the user is interested in. Therefore, the control unit 10 and the search device 1 in the present embodiment can appropriately provide useful information to the user.

Moreover, in this embodiment, the search control part 12 is a topic according to a predetermined word, when the word extracted by the speech recognition part 11 contains the predetermined word (prescribed word) defined beforehand. Switch between provision mode and conversation tracking mode.
Thereby, since the two search modes of the topic providing mode and the conversation tracking mode can be switched appropriately, the control unit 10 and the search device 1 in the present embodiment appropriately provide useful information to the user. Can do.

Moreover, in this embodiment, the search control part 12 produces | generates the search expression which added the keyword (search word) of the word extracted by the speech recognition part 11 sequentially as a search condition, and based on the produced | generated search expression. Thus, the search server 4 is made to search for information such as images.
Thereby, it becomes possible to widen the search condition and narrow the search condition by sequentially inputting the voice information. Therefore, the control unit 10 and the search device 1 in the present embodiment can make the search server 4 appropriately search in the topic providing mode and the conversation tracking mode.

In the present embodiment, the search control unit 12 combines keywords by logical sum (OR) in the topic providing mode and adds them to the search formula, and combines keywords by logical product (AND) in the conversation tracking mode. To add to the search expression.
Thereby, the control part 10 and the search device 1 in this embodiment can generate | occur | produce the search formula by two search modes appropriately with a simple means.

In the present embodiment, the search control unit 12 excludes keywords from the search formula in the oldest order extracted by the speech recognition unit 11.
Thereby, since the control part 10 and the search device 1 in this embodiment delete an old keyword from a search formula, it can be made to search with respect to the search server 4 by a suitable search condition.

In the present embodiment, the search control unit 12 excludes a keyword from the search formula when the number of search results is equal to or greater than a first predetermined number in the topic providing mode.
Thereby, the number of search results is reduced. Therefore, the control unit 10 and the search device 1 in the present embodiment can reduce a situation where information such as an image that is a search result is excessive and a user is provided with an unnecessary search result. it can.

In the present embodiment, the search control unit 12 excludes the keyword from the search formula when the number of search results is equal to or less than a predetermined second specified number in the conversation tracking mode.
This increases the number of search results. Therefore, the control unit 10 and the search device 1 according to the present embodiment can reduce the situation in which the user cannot provide the user with the necessary search results due to excessive narrowing down.

In the present embodiment, the voice recognition unit 11 recognizes input voice information and converts it into text information, and extracts a word included in the text information.
Thereby, the control part 10 and the search device 1 in this embodiment extract a word from the audio | voice information input from the sound collection microphone 2, and information, such as an image (2nd information) by the search conditions based on the extracted word. Can be searched. Therefore, the control unit 10 and the search device 1 according to the present embodiment can search and provide appropriate information (second information) from natural conversations that the user is not conscious of.

In addition, the search device 1 (output device) in the present embodiment includes a control unit 10 and a display unit 20 that outputs information such as an image searched by the control unit 10.
Thereby, the search device 1 in this embodiment can appropriately provide information such as searched images to the user.

In the above-described embodiment, an example has been described in which the search control unit 12 switches between the topic providing mode and the conversation tracking mode depending on whether or not the specified word (predetermined word) is included. It is also possible to switch between these two modes.
Next, a modified example for switching between the two modes will be described.

<First Modification>
In the first modification, the search control unit 12 switches between the topic provision mode and the conversation tracking mode according to a predetermined time. Here, the predetermined time is, for example, two-thirds (2/3) of the scheduled conference period, a time ten minutes before the scheduled end time, a predetermined elapsed time from the start of the conference, or the like.

FIG. 5 is a flowchart showing a first modification of the topic providing mode in the present embodiment.
The processing in the topic providing mode in this figure is the same as the processing in the topic providing mode shown in FIG. 3 except that the processing in step S105a is different. Here, the process of step S105a will be described.

In step S105a, the search control unit 12 of the control unit 10 determines whether or not a predetermined time has elapsed. Here, as described above, the predetermined time is, for example, two-thirds (2/3) of the scheduled conference period. When it is determined that the predetermined time has elapsed (step S105a: YES), the search control unit 12 shifts the processing from the topic providing mode to the conversation tracking mode. If the search control unit 12 determines that the predetermined time has not elapsed (step S105a: NO), the search control unit 12 advances the process to step S106.
Although illustration is omitted, in the conversation tracking mode, the conversation tracking mode may be switched to the topic providing mode in accordance with a predetermined time as in FIG.

As described above, the search control unit 12 switches between the topic provision mode and the conversation tracking mode according to a predetermined time.
Thereby, since the two search modes of the topic providing mode and the conversation tracking mode can be switched appropriately, the control unit 10 and the search device 1 in the present embodiment appropriately provide useful information to the user. Can do. For example, in the conference, the topic providing mode and the conversation tracking mode are switched according to a predetermined time, so that the control unit 10 and the search device 1 in the present embodiment provide useful information according to the progress of the conference. Can do.

<Second Modification>
In the second modified example, the search control unit 12 switches between the topic providing mode and the conversation tracking mode when the keyword extracted by the speech recognition unit 11 includes a keyword included in the search expression. That is, the search control unit 12 switches between the topic provision mode and the conversation tracking mode when the same keyword (search word) appears in the voice information (for example, voice information in conversation).

FIG. 6 is a flowchart showing a first modification of the topic providing mode in the present embodiment.
The processing in the topic providing mode in this figure is the same as the processing in the topic providing mode shown in FIG. 3 except that the processing in step S105b is different. Here, the process of step S105b will be described.

  In step S105b, the control unit 10 determines whether or not the same speech word (keyword) is included in the input voice information. That is, the search control unit 12 of the control unit 10 determines whether or not a keyword included in the search formula is included in the words extracted by the speech recognition unit 11. When it is determined that the extracted word includes a keyword included in the search expression (step S105b: YES), the search control unit 12 shifts the processing from the topic providing mode to the conversation tracking mode. If the search control unit 12 determines that the extracted word does not include the keyword included in the search expression (step S105b: NO), the search control unit 12 advances the process to step S106.

As described above, the search control unit 12 switches between the topic providing mode and the conversation tracking mode when the keyword extracted by the speech recognition unit 11 includes a keyword included in the search expression.
Thereby, since the two search modes of the topic providing mode and the conversation tracking mode can be switched appropriately, the control unit 10 and the search device 1 in the present embodiment appropriately provide useful information to the user. Can do.

<Third Modification>
In the third modification, the search control unit 12 switches between the topic providing mode and the conversation tracking mode when a specific speaker is speaking.
For example, the voice recognition unit 11 identifies a speaker based on the voice information input from the sound collection microphone 2, associates the information indicating the speaker with the extracted word, and supplies the information to the search control unit 12. Note that the voice recognition unit 11 may identify the speaker based on the feature amount in the voice information such as a voiceprint, or identify the speaker from the direction of the voice based on the voice information input from the plurality of microphones 2. May be.
For example, when the speaker is a person who brings together the meeting like the chairperson in the meeting, the search control unit 12 determines that the meeting has shifted to the summary, and switches from the topic providing mode to the conversation tracking mode. The search control unit 12 switches the topic providing mode from the conversation tracking mode when, for example, the speaker is a person who describes a reference opinion or a person who presents various opinions.

In this way, the search control unit 12 switches between the topic providing mode and the conversation tracking mode when a specific speaker is speaking.
Thereby, since the two search modes of the topic providing mode and the conversation tracking mode can be switched appropriately, the control unit 10 and the search device 1 in the present embodiment appropriately provide useful information to the user. Can do.

  In the present embodiment, when switching between the topic providing mode and the conversation tracking mode, the search control unit 12 deletes the search expression once, for example, using the most recent keyword or the most recent predetermined number of keywords, A new search expression may be generated. For example, when the topic providing mode is switched to the conversation tracking mode, the search control unit 12 may generate a search expression by combining the most recent predetermined number of keywords by AND (AND). For example, when the conversation tracking mode is switched to the topic providing mode, the search control unit 12 may generate a search expression by combining the most recent predetermined number of keywords with a logical sum (OR). Good.

Further, when switching between the topic providing mode and the conversation tracking mode, the search control unit 12 may change the operator of only the most recent predetermined number of keywords. For example, when the topic providing mode is switched to the conversation tracking mode, the search control unit 12 combines the most recent predetermined number of keywords with a logical product (AND), and the rest remains a logical sum (OR). The search expression may be generated by. Further, for example, when the conversation tracking mode is switched to the topic providing mode, the search control unit 12 combines the most recent predetermined number of keywords with a logical sum (OR), and the remainder is logically ANDed. The search expression may be generated as it is.
Further, when switching between the topic providing mode and the conversation tracking mode, the search control unit 12 may maintain the search formula as it is, and generate the search formula by changing the operator from the newly added keyword.

Next, a search control device and a search device (output device) according to a second embodiment will be described with reference to the drawings.
[Second Embodiment]
The system configuration of the search device 1 in the second embodiment is the same as the system configuration of the search device 1 in the first embodiment shown in FIG.
FIG. 7 is a block diagram of the search device 1 according to the second embodiment.
In FIG. 7, the search device 1 (output device) includes a control unit 10 a and a display unit 20. In the present embodiment, the control unit 10a functions as a search control device.
In this figure, the same components as those in FIG. 2 are denoted by the same reference numerals, and the description thereof is omitted.

The control unit 10 a controls each unit of the search device 1. Based on the audio information (first information) input from the sound collection microphone 2, the control unit 10a causes the search server 4 to search via the network 3, and acquires information such as an image as a search result. In addition, the control unit 10 a causes the display unit 20 to display information such as an image acquired from the search server 4. In the present embodiment, when the input voice information (sentence) is a negative sentence, the control unit 10a determines that the user does not want the word included in the sentence, and starts from the previous keyword. Search again without the words contained in the negative sentence.
In addition, the control unit 10 a includes a voice recognition unit 11, a search control unit 12, and a negative sentence detection unit 13.

  The negative sentence detection unit 13 (detection unit) detects a negative sentence included in the voice information input from the sound collection microphone 2. Here, the negative sentence is a sentence including a word indicating negative. Also, the words indicating negative are “not”, “not”, “no”, “no”, and the like. For example, the negative sentence detection unit 13 acquires, from the speech recognition unit 11, the text information that the speech recognition unit 11 has recognized the speech information and the words extracted by the speech recognition unit 11. The negative sentence detection unit 13 detects a negative sentence based on the text information acquired from the voice recognition unit 11 and the word. For example, the negative sentence detection unit 13 extracts a sentence based on text information, and detects a negative sentence based on whether or not a word indicating negative is included in the extracted sentence. The negative sentence detection unit 13 supplies the detected negative sentence and the words included in the negative sentence to the search control unit 12.

When a negative sentence is detected by the negative sentence detection unit 13, the search control unit 12 excludes a keyword included in the negative sentence from the search formula. For example, the search control unit 12 determines whether or not there is a negative sentence supplied from the negative sentence detection unit 13. Furthermore, when it is determined that there is a negative sentence, the search control unit 12 determines whether or not the word supplied from the negative sentence detection unit 13 is in the search formula. Then, the search control unit 12 deletes the word determined to be in the search formula from the search formula. In addition, when the search control unit 12 determines that there is a negative sentence, the keyword included in the newly input voice information (first information) is logically ORed according to two search modes. Combined by AND (AND) and added to the search expression.
Since other processing in the search control unit 12 is the same as that in the first embodiment, description thereof is omitted here.

Next, the operation of the search device 1 in this embodiment will be described.
FIG. 8 is a flowchart showing an example of processing in the topic provision mode in the present embodiment.
In this figure, in this embodiment, the point which the control part 10a performs the determination process by a negative sentence differs from the process of the topic provision mode in 1st Embodiment shown by FIG.
In this figure, first, the control unit 10a determines whether or not there is an audio input from the sound collection microphone 2 (step S201). That is, the control unit 10 a determines whether or not audio information is input from the sound collection microphone 2 to the audio recognition unit 11. When it is determined that the voice information has been input (step S201: YES), the control unit 10a advances the process to step S202. When the control unit 10a determines that no audio information is input (step S201: NO), the control unit 10a returns the process to step S201, and repeats the process of step S201 until the audio information is input.

  Next, the control unit 10a performs voice recognition and morphological analysis (step S202). That is, the voice recognition unit 11 of the control unit 10a recognizes input voice information, converts it into text information (sentence), and extracts words included in the text information. The voice recognition unit 11 supplies the extracted word to the search control unit 12. The voice recognition unit 11 also supplies the converted text information (sentence) and the extracted word to the negative sentence detection unit 13.

  Next, the control unit 10a determines whether or not there is a negative sentence (step S203). First, the negative sentence detection part 13 of the control part 10a detects a negative sentence based on the text information acquired from the speech recognition part 11, and a word. The negative sentence detection unit 13 supplies the detected negative sentence and the words included in the negative sentence to the search control unit 12. The search control unit 12 determines whether there is a negative sentence supplied from the negative sentence detection unit 13. When it is determined that there is a negative sentence (step S203: YES), the search control unit 12 advances the process to step S204. If the search control unit 12 determines that there is no negative sentence (step S203: NO), the search control unit 12 advances the process to step S205.

  Next, in step S204, the control unit 10a executes search word (keyword) deletion and search processing. That is, the search control unit 12 of the control unit 10a deletes the word determined to be in the search formula from the search formula. The search control unit 12 supplies the generated search expression to the search server 4 via the network 3, and acquires information such as images as search results from the search server 4 via the network 3.

  In step S205, the control unit 10a executes search word (keyword) addition and search processing. In other words, the search control unit 12 of the control unit 10a combines the keywords included in the newly input voice information (first information) with a logical sum (OR) and adds them to the search formula. The search control unit 12 supplies the generated search expression to the search server 4 via the network 3, and acquires information such as images as search results from the search server 4 via the network 3.

The processing from the next step S206 to step S209 is the same as the processing from step S104 to step S107 shown in FIG. 3, and a description thereof will be omitted here.
Moreover, although illustration is omitted, also in the process of the conversation tracking mode, the control unit 10a executes a determination process using a negative sentence as in Step S203 and Step S204 of FIG.

As described above, the control unit 10a according to the present embodiment detects a negative sentence that is a negative sentence that is a sentence including a word indicating negative and is included in the voice information (first information). Part 13 is provided. When a negative sentence is detected by the negative sentence detection unit 13, the search control unit 12 excludes a keyword included in the negative sentence from the search formula.
Thereby, the control part 10a and the search device 1 in this embodiment can remove a keyword (search word) from a search expression. For example, the control unit 10a and the search device 1 in the present embodiment can remove an incorrect keyword (search word) from the previous search formula when a recognition error by the voice recognition unit 11 occurs. Therefore, since the control part 10a and the search device 1 in this embodiment can perform an appropriate search for the user, it is possible to appropriately provide useful information to the user.

Next, a search control device and a search device (output device) according to a third embodiment will be described with reference to the drawings.
[Third Embodiment]
FIG. 9 is a second diagram illustrating an example of the configuration of the search device 1 according to the present embodiment.
In FIG. 9, the search device 1 is connected to a search server 4 and a synonym search server 5 via a network 3 (for example, the Internet). In addition, the search device 1 is connected to a sound collection microphone 2. Thereby, voice information (for example, voice signal of conversation) of the user (for example, two users (U 1, U 2)) is input from the sound collection microphone 2 to the search device 1.
In this figure, the same components as those in FIG. 2 are denoted by the same reference numerals, and the description thereof is omitted. Note that the present embodiment is different from the first and second embodiments described above in that it is connected to the synonym search server 5.

  The synonym search server 5 searches for synonyms that are related words related to the input word based on the word input via the network 3, and outputs the searched synonym via the network 3. The synonyms are, for example, “Apple”, “Apple” and the like for the word “apple”. For example, the synonym search server 5 may search for synonyms via the network 3 or may search for synonyms from information stored in a database (for example, a synonym dictionary) provided in the own device.

FIG. 10 is a block diagram of the search device 1 according to the third embodiment.
In FIG. 10, the search device 1 (output device) includes a control unit 10 b and a display unit 20. In the present embodiment, the control unit 10b functions as a search control device.
In this figure, the same components as those in FIG. 7 are denoted by the same reference numerals, and description thereof is omitted.

The control unit 10b controls each unit of the search device 1. The control unit 10b causes the search server 4 to search through the network 3 based on the voice information (first information) input from the sound collection microphone 2, and acquires information such as an image as a search result. In addition, the control unit 10 b causes the display unit 20 to display information such as an image acquired from the search server 4. Moreover, in this embodiment, the control part 10b acquires the synonym relevant to the word contained in audio | voice information (1st information) from the synonym search server 5, Based on the acquired synonym, an image etc. are used as a search result. Get information.
In addition, the control unit 10 b includes a voice recognition unit 11, a search control unit 12, a negative sentence detection unit 13, and a synonym acquisition unit 14.

  The synonym acquisition unit 14 (acquisition unit) acquires a synonym (related word) related to the word based on the word extracted by the speech recognition unit 11. That is, the synonym acquisition unit 14 supplies the word supplied from the search control unit 12 to the synonym search server 5 via the network 3 and causes the synonym search server 5 to search for synonyms. Then, the synonym acquisition unit 14 supplies the synonym supplied from the synonym search server 5 via the network 3 to the search control unit 12.

The search control unit 12 acquires a synonym from the synonym acquisition unit 14 based on the word extracted by the speech recognition unit 11, and based on a search condition based on the acquired synonym and the word extracted by the speech recognition unit 11. The search server 4 is made to search for information such as images. For example, the search control unit 12 first causes the search server 4 to search for information such as an image based on a search condition based on the word extracted by the voice recognition unit 11. Then, the search control unit 12 is extracted by the synonym and speech recognition unit 11 when the number of search results based on the words extracted by the speech recognition unit 11 is equal to or less than a predetermined third specified number. Based on the search condition based on the word, the search server 4 is caused to search for information such as an image.
Since other processing in the search control unit 12 is the same as that in the second embodiment, description thereof is omitted here.

Next, the operation of the search device 1 in this embodiment will be described.
FIG. 11 is a flowchart showing an example of processing in the topic provision mode in the present embodiment.
In this figure, in this embodiment, the point that the control part 10b performs the search process based on a synonym differs from the process of the topic provision mode in 2nd Embodiment shown by FIG.

In this figure, the processing from step S201 to step S207 is the same as that of the second embodiment shown in FIG. Here, the processing after step S208a will be described.
In step S208a, the control unit 10b determines whether or not the search result is equal to or greater than a specified number (first specified number). That is, the search control unit 12 of the control unit 10b determines whether or not the number of search results acquired from the search server 4 is equal to or greater than a specified number (first specified number). If the search control unit 12 determines that the number of search results is equal to or greater than the specified number (first specified number) (step S208a: YES), the search control unit 12 advances the process to step S209. If the search control unit 12 determines that the number of search results is less than the specified number (first specified number) (step S208a: NO), the search control unit 12 advances the process to step S210.

  Since the process of step S209 is the same as that of 2nd Embodiment shown by FIG. 8, description is abbreviate | omitted here.

  In step S210, the control unit 10b determines whether or not the search result is equal to or less than a specified number (third specified number). That is, the search control unit 12 of the control unit 10b determines whether or not the number of search results acquired from the search server 4 is equal to or less than a specified number (third specified number). If the search control unit 12 determines that the number of search results is equal to or greater than the specified number (third specified number) (step S210: YES), the search control unit 12 advances the process to step S211. If the search control unit 12 determines that the number of search results exceeds the specified number (third specified number) (step S210: NO), the search control unit 12 returns the process to step S201.

Next, in step S211, the control unit 10b searches for synonyms and adds the synonyms as keywords. That is, the search control unit 12 acquires a synonym from the synonym acquisition unit 14 based on the keyword (search word) extracted by the speech recognition unit 11. The search control unit 12 adds the acquired synonym to the search expression to generate a new search expression, and returns the process to step S201. As a result, the search control unit 12 causes the search server 4 to search for information such as an image based on the newly generated search expression.
In addition, when adding the acquired synonym to a search formula in step S211, the search control part 12 couple | bonds a keyword and the corresponding synonym by OR, for example, and adds it to a search formula.

  Although not shown, also in the conversation tracking mode process, the control unit 10b executes a process of adding a synonym to the search expression, similarly to steps S210 and S211 of FIG. In addition, in the process of the conversation tracking mode, when adding the acquired synonym to the search expression, for example, the search control unit 12 combines the keyword and the corresponding synonym by a logical sum (OR) and adds it to the search expression. .

As described above, the control unit 10b according to the present embodiment includes the synonym acquisition unit 14 that acquires a synonym (related word) related to a word based on the word extracted by the voice recognition unit 11. The search control unit 12 acquires a synonym from the synonym acquisition unit 14 based on the word extracted by the speech recognition unit 11, and based on a search condition based on the acquired synonym and the word extracted by the speech recognition unit 11. Thus, the search server 4 is made to search for information such as images.
Thereby, information such as an image can be searched based on a search condition including a synonym, so that the control unit 10b and the search device 1 in the present embodiment can perform a flexible search. Therefore, since the control part 10b and the search device 1 in this embodiment can perform an appropriate search for the user, it is possible to appropriately provide useful information to the user.

In the present embodiment, the search control unit 12 has the number of search results based on the words (search words) extracted by the speech recognition unit 11 equal to or less than a predetermined value (third specified number). In this case, the search server 4 is made to search for information such as images based on search conditions based on synonyms and words (search words).
Thereby, the control part 10b and the search device 1 in this embodiment can adjust the number of search results flexibly. Therefore, the control unit 10b and the search device 1 in the present embodiment can appropriately provide useful information to the user.

According to the present embodiment, the control unit 10b (detection control device) includes the voice recognition unit 11 and the search control unit 12. The voice recognition unit 11 extracts words included in the input voice information. The search control unit 12 causes the search server 4 to search for information such as an image output as a search result based on the words extracted by the voice recognition unit 11. Then, when the number of search results based on the word is equal to or smaller than a predetermined number, the search control unit 12 determines information such as an image based on a synonym (related word) related to the word. To the search server 4.
Thereby, the control part 10b and the search device 1 in this embodiment can adjust the number of search results flexibly. Therefore, the control unit 10b and the search device 1 in the present embodiment can appropriately provide useful information to the user.

Next, a search control device and a search device (output device) according to a fourth embodiment will be described with reference to the drawings.
[Fourth Embodiment]
The system configuration of the search device 1 in the fourth embodiment is the same as the system configuration of the search device 1 in the third embodiment shown in FIG.
The configuration of the search device 1 is the same as the system configuration of the search device 1 in the third embodiment shown in FIG.

In the present embodiment, the search control unit 12 searches for a keyword excluded from the search formula when the word extracted by the speech recognition unit 11 includes a word (resurrection word) indicating a predetermined revival. Add to. Here, the word indicating the predetermined resurrection is, for example, a word such as “Sakaki no Yatsu” or “A little before”. For example, the search control unit 12 determines whether the word extracted by the voice recognition unit 11 includes the above-described word indicating resurrection. Further, when the search control unit 12 determines that the above-described word indicating resurrection is included, the search control unit 12 adds a keyword that has been excluded from the search formula in the past to the search formula. In this case, for example, the search control unit 12 adds the new one of the excluded keywords to the search formula and restores it. When restoring the excluded keywords, the search control unit 12 adds the keywords to be restored to the search formula by combining them with logical sum (OR) or logical product (AND) according to two search modes. .
Since other processing in the search control unit 12 is the same as in the third embodiment, description thereof is omitted here.

Next, the operation of the search device 1 in this embodiment will be described.
FIG. 12 is a flowchart illustrating an example of processing in the topic provision mode in the present embodiment.
In this figure, in this embodiment, the point which the control part 10b performs the determination process of whether a word contains a resurrection word differs from the process of the topic provision mode in 3rd Embodiment shown by FIG.
In this figure, first, the control unit 10b determines whether or not there is an audio input from the sound collection microphone 2 (step S301). That is, the control unit 10 b determines whether or not voice information is input from the sound collection microphone 2 to the voice recognition unit 11. When it is determined that the voice information has been input (step S301: YES), the control unit 10b advances the process to step S302. When it is determined that the voice information is not input (step S301: NO), the control unit 10b returns the process to step S301, and repeats the process of step S301 until the voice information is input.

  Next, the control unit 10b performs voice recognition and morphological analysis (step S302). That is, the voice recognition unit 11 of the control unit 10b recognizes the input voice information, converts it into text information (sentence), and extracts words included in the text information. The voice recognition unit 11 supplies the extracted word to the search control unit 12. The voice recognition unit 11 also supplies the converted text information (sentence) and the extracted word to the negative sentence detection unit 13.

  Next, the control unit 10b determines whether or not the word includes a resurrection word (step S303). First, the search control unit 12 of the control unit 10b determines whether or not the word acquired from the speech recognition unit 11 includes a resurrection word that is a word indicating the above-described predetermined resurrection. When it is determined that the resurrection word is included (step S303: YES), the search control unit 12 advances the process to step S304. If the search control unit 12 determines that the resurrection word is not included (step S303: NO), the process proceeds to step S305.

  Next, in step S304, the control unit 10b executes search word (keyword) restoration and search processing. That is, for example, the search control unit 12 of the control unit 10b restores the search formula in order from the newest one of the keywords excluded from the search formula in the past. The search control unit 12 supplies the generated search expression to the search server 4 via the network 3, and acquires information such as images as search results from the search server 4 via the network 3. Thereafter, the search control unit 12 advances the process to step S308.

The processing from the next step S305 to step S313 is the same as the processing from step S206 to step S211 shown in FIG. 11, and a description thereof will be omitted here.
Although illustration is omitted, also in the conversation tracking mode process, the control unit 10b performs a process of determining whether or not a word includes a resurrection word, as in Steps S303 and S304 of FIG.

As described above, the search control unit 12 of the control unit 10b according to the present embodiment uses the search formula when the word extracted by the speech recognition unit 11 includes a word indicating a predetermined resurrection. Add the excluded keyword to the search expression.
As a result, keywords (search words) that have been accidentally deleted (excluded) from the search formula can be restored, so that the control unit 10b and the search device 1 in this embodiment can appropriately perform search processing. Since the control part 10b and the search device 1 in this embodiment can perform an appropriate search for the user, it is possible to appropriately provide useful information to the user.

Next, a search control device and a search device (output device) according to a fifth embodiment will be described with reference to the drawings.
[Fifth Embodiment]
FIG. 13 is a block diagram of the search device 1 according to the fifth embodiment.
In FIG. 13, the search device 1 (output device) includes a control unit 10b, a display unit 20, a search unit 4a, and a synonym search unit 5a. In the present embodiment, the control unit 10b, the search unit 4a, and the synonym search unit 5a function as the search control device 30.
In this figure, the same components as those in FIG. 10 are denoted by the same reference numerals, and the description thereof is omitted.

The search control device 30 and the search device 1 in the present embodiment are different from the fourth embodiment in that they include a search unit 4a and a synonym search unit 5a.
The search unit 4a (search means) has the same function as the search server 4 in the fourth embodiment, and based on a keyword (search word) or a search expression based on the keyword, an image, sound, and text related to the keyword Search for information such as For example, the search unit 4a may search for information as a search result via a network, or search for information as a search result from information stored in an internal database (not shown). Good.

  The synonym search unit 5a has a function similar to that of the synonym search server 5 in the fourth embodiment. Based on the word supplied from the synonym acquisition unit 14, the synonym search unit 5a selects a synonym that is a related word related to the supplied word. The retrieved synonym is output to the synonym acquisition unit 14. For example, the synonym search unit 5a may search for synonyms via a network, or may search for synonyms from information stored in a database (not shown) (for example, a synonym dictionary) provided inside.

  Note that the search control device 30 and the search device 1 in the present embodiment are the same as the operations in the fourth embodiment, and are omitted here.

As described above, the search control device 30 according to this embodiment includes the search unit 4a (search means) that searches for information such as images.
Thereby, the search control device 30 according to the present embodiment can appropriately provide useful information to the user even in an environment (offline) where connection to the network is not possible.

The present invention is not limited to the above embodiments, and can be modified without departing from the spirit of the present invention.
For example, in each of the above-described embodiments, the form in which audio information is used as the first information has been described, but the present invention is not limited to this. For example, the embodiments described above can be applied to communication using text information such as chat or email. In this case, since the voice recognition function is unnecessary, each of the above embodiments can be realized with a simple configuration.

  Further, in each of the embodiments described above, the form of searching for information such as an image has been described as an example of a search result. Further, in each of the above embodiments, the form in which the search device 1 includes the display unit 20 as an output unit has been described. However, the output unit may be an output device such as a speaker that outputs sound information, or may be in another form. . Further, the search device 1 may include an interface that outputs the search result as an output unit, and may output the display unit 20 or a speaker connected to the outside via the interface.

  Further, in each of the above-described embodiments, the form in which a plurality of users such as a conference use the search device 1 has been described. However, the search device 1 may be used by one user. As described above, the search device 1 can appropriately provide useful information even when one person examines or summarizes something.

Further, in each of the above embodiments, the form using logical sum and logical product as the logical operators for generating the search expressions has been described, but other logical operators may be used.
Moreover, the combination with each embodiment or each modification in each said embodiment is not limited above. For example, the third embodiment has been described in conjunction with the second embodiment, but may be implemented in combination with the first embodiment. Similarly, the fourth embodiment has been described in conjunction with the third embodiment, but may be implemented in combination with the first or second embodiment. Similarly, the fifth embodiment has been described in conjunction with the third or fourth embodiment, but may be implemented in combination with the first or second embodiment.

In addition, each part with which the control part 10 (10a, 10b) and the search control apparatus 30 in each said embodiment is provided may be implement | achieved by exclusive hardware, and the control part 10 (10a, 10b). ) And the search control device 30 are configured by a memory and a CPU (central processing unit), and a program for realizing the functions of the control unit 10 (10a, 10b) and the search control device 30 is loaded into the memory. The function may be realized by executing as described above.
Moreover, the above-mentioned control part 10 (10a, 10b), the search control apparatus 30, and the search apparatus 1 have a computer system inside. The processing steps (processing procedures) of the control unit 10 (10a, 10b), the search control device 30, and the search device 1 described above are stored in a computer-readable recording medium in the form of a program. The above processing may be performed by reading and executing.

  Further, a program for realizing the functions of the respective units included in the control unit 10 (10a, 10b), the search control device 30, and the search device 1 is recorded on a computer-readable recording medium, and the program recorded on the recording medium May be processed by the control unit 10 (10a, 10b), the search control device 30 and the respective units included in the search device 1 by executing the above-described information into a computer system. Here, the “computer system” includes an OS and hardware such as peripheral devices.

Further, the “computer system” includes a homepage providing environment (or display environment) if a WWW system is used.
The “computer-readable recording medium” refers to a storage device such as a flexible medium, a magneto-optical disk, a portable medium such as a ROM and a CD-ROM, and a hard disk incorporated in a computer system. Furthermore, the “computer-readable recording medium” dynamically holds a program for a short time like a communication line when transmitting a program via a network such as the Internet or a communication line such as a telephone line. In this case, a volatile memory in a computer system serving as a server or a client in that case, and a program that holds a program for a certain period of time are also included. The program may be a program for realizing a part of the functions described above, and may be a program capable of realizing the functions described above in combination with a program already recorded in a computer system.

  DESCRIPTION OF SYMBOLS 1 ... Search apparatus, 4 ... Search server, 4a ... Search part, 5 ... Synonym search server, 5a ... Synonym search part, 10, 10a, 10b ... Control part (search control apparatus), 11 ... Voice recognition part, 12 ... Search Control part, 13 ... Negative sentence detection part, 14 ... Synonym acquisition part, 20 ... Display part, 30 ... Search control apparatus

Claims (18)

An extraction unit that extracts words included in the input first information;
A control unit that causes a search unit to search for second information output as a search result based on a search condition based on the word extracted by the extraction unit;
With
The controller is
A search control apparatus comprising: a first search mode for widening the search condition; and a second search mode for narrowing the search condition. The controller is
Switching between the first search mode and the second search mode according to the predetermined word when the predetermined word is included in the word extracted by the extraction unit. The search control device according to claim 1. The controller is
The search control device according to claim 1 or 2, wherein the first search mode and the second search mode are switched according to a predetermined time. The controller is
The search formula in which keywords of the words are sequentially added is generated as the search condition, and the second information is searched by the search unit based on the generated search formula. The search control device according to any one of claims 1 to 3. The controller is
The first search mode and the second search mode are switched when the keyword extracted by the extraction unit includes the keyword included in the search formula. The search control device described in 1. A negative sentence that is a sentence including a word indicating negative, and includes a detection unit that detects a negative sentence included in the first information;
The controller is
The search control device according to claim 4 or 5, wherein, when the negative sentence is detected by the detection unit, the keyword included in the negative sentence is excluded from the search formula. The controller is
In the first search mode, the keywords are combined by logical sum and added to the search formula,
The search control device according to any one of claims 4 to 6, wherein in the second search mode, the keywords are combined by logical product and added to the search formula. The controller is
The search control device according to any one of claims 4 to 7, wherein the keyword is excluded from the search formula in the oldest order extracted by the extraction unit. The controller is
The said keyword is excluded from the said search formula, when the number of the said search results becomes more than the predetermined 1st predetermined number in the said 1st search mode. Search control device. The controller is
The said keyword is excluded from the said search formula, when the number of the said search results becomes below a predetermined 2nd predetermined number in the said 2nd search mode. 9. The search control device according to 9. The controller is
5. The keyword excluded from the search formula is added to the search formula when the word extracted by the extraction unit includes a word indicating a predetermined resurrection. The search control device according to claim 10. Based on the word extracted by the extraction unit, an acquisition unit that acquires a related word related to the word,
The controller is
Based on the word extracted by the extraction unit, the related word is acquired from the acquisition unit, and the second information is acquired based on a search condition based on the acquired related word and the word. The search control device according to any one of claims 1 to 11, wherein the search is performed. The controller is
When the number of search results based on the word is equal to or less than a predetermined third predetermined number, the search means is configured to search the second information based on a search condition based on the related word and the word. The search control device according to claim 12, wherein the search control device is made to search.   The search control device according to claim 1, further comprising the search unit that searches for the second information. The first information is audio information;
The extraction unit includes:
The search control according to any one of claims 1 to 14, wherein the input voice information is recognized and converted into text information, and the word included in the text information is extracted. apparatus. An extraction unit that extracts words included in the input first information;
A control unit that causes a search unit to search for second information output as a search result based on the word extracted by the extraction unit;
With
The controller is
When the number of the search results based on the word is equal to or less than a predetermined number, the search unit is made to search the second information based on a related word related to the word. A featured search control device. The search control device according to any one of claims 1 to 16,
An output device comprising: an output unit that outputs the second information searched by the search control device. A program for causing a computer as a search control device to execute,
An extraction procedure in which the extraction unit extracts words included in the input first information;
A control procedure for causing the search means to search for second information output as a search result based on a search condition based on the word extracted by the extraction procedure;
Including
The control procedure is:
A program comprising: a first search mode for widening the search condition; and a second search mode for narrowing the search condition.

Images