JP2013238986A - Processing apparatus, processing system, and output method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a processing apparatus, a processing system, and an output method and program for achieving more natural responses.SOLUTION: A processing apparatus includes: a voice recognition unit 21 that recognizes a user's voice; a search request unit 25 that requests search for information on the basis of the voice recognized by the voice recognition unit 21; an acquisition unit 27 that acquires a search result on the basis of the search for information requested by the search request unit 25; a situation recognition unit 29 that recognizes a situation of the user; and an output control unit 33 that determines whether or not it is a timing for causing an output unit 19 to output the search result on the basis of the situation of the user recognized by the situation recognition unit 29, and if it is the timing for the output, causes the output unit 19 to output the search result.

Description

  The present invention relates to a processing apparatus, a processing system, an output method, and a program.

  2. Description of the Related Art Conventionally, a technique for causing a computer to recognize a voice uttered by a user, understand and infer the intention of the user's utterance, and perform an appropriate response is known. In such a technique, it is required to make a computer perform a natural response that allows humans to interact with each other. For example, Patent Document 1 discloses a technology that responds to an answer to a question when the user asks a question.

  However, in the related art as described above, since the timing of response is not considered, the response is not always performed at an appropriate timing for the user. In particular, when a dialogue is performed between a plurality of users, there is a possibility that the dialogue between the users is hindered.

  The present invention has been made in view of the above circumstances, and an object thereof is to provide a processing device, a processing system, an output method, and a program that can cause a more natural response.

  In order to solve the above-described problems and achieve the object, a processing device according to an aspect of the present invention provides a voice recognition unit that recognizes a user's voice and information based on the voice recognized by the voice recognition unit. A search request unit that requests a search, an acquisition unit that acquires a search result based on a search for information requested by the search request unit, a situation recognition unit that recognizes the user's situation, and the situation recognition unit An output control unit that determines whether or not it is a timing to output the search result to the output unit based on the user status recognized in step S3, and outputs the search result to the output unit if it is a timing to output the search result And comprising.

  A processing system according to another aspect of the present invention includes a speech recognition unit that recognizes a user's voice, a search request unit that requests a search for information based on the speech recognized by the speech recognition unit, A search unit that searches for information requested by a search request unit, an acquisition unit that acquires a search result of information of the search unit, a situation recognition unit that recognizes the user's situation, and a situation recognition unit that is recognized An output control unit that determines whether it is a timing to output the search result to an output unit based on a user's situation, and outputs the search result to the output unit when it is a timing to output the search result; .

  An output method according to another aspect of the present invention includes: a voice recognition step in which a voice recognition unit recognizes a user's voice; and a search request unit based on the voice recognized by the voice recognition unit. A search requesting step for requesting a search, an acquisition unit for acquiring a search result based on a search for information requested by the search requesting unit, and a situation in which the situation recognition unit recognizes the situation of the user When the recognition step and the output control unit determine whether it is a timing to output the search result to the output unit based on the user situation recognized by the situation recognition unit, And an output control step for causing the output unit to output the search result.

  According to another aspect of the present invention, there is provided a program for recognizing a user's voice, a search requesting step for requesting information search based on the voice recognized by the voice recognition unit, and the search Based on a search for information requested by the request unit, an acquisition step for acquiring a search result, a situation recognition step for recognizing the situation of the user, and a situation of the user recognized by the situation recognition unit, It is determined whether or not it is time to output the search result to the output unit, and if it is time to output, the output control step for outputting the search result to the output unit is for causing the computer to function. is there.

  According to the present invention, there is an effect that a more natural response can be performed.

FIG. 1 is a block diagram illustrating an example of a configuration of a processing system according to the present embodiment. FIG. 2 is a block diagram illustrating an example of the configuration of the voice recognition unit of the present embodiment. FIG. 3 is a flowchart showing an example of processing executed in the processing system of the present embodiment.

  Hereinafter, embodiments of a processing device, a processing system, an output method, and a program according to the present invention will be described in detail with reference to the accompanying drawings.

  First, the configuration of the processing system of this embodiment will be described.

  FIG. 1 is a block diagram showing an example of the configuration of the processing system 1 of the present embodiment. As illustrated in FIG. 1, the processing system 1 includes a network agent (hereinafter referred to as “NA”) 10 as an example of a processing device, and a search server 101 as an example of a search unit. The NA 10 and the search server 101 are connected via the Internet 107.

  The search server 101 searches information published on the Web, and may be anything that provides a search engine function on the Web, for example. Specifically, the search server 101 receives a search query from the NA 10, searches information published on the Web in accordance with the received search query, and transmits the search result to the NA 10. Here, the information searched by the search server 101 may be dynamic information published on a Web dynamic page or static information published on a Web static page. . In the example illustrated in FIG. 1, one search server is illustrated, but the number is not limited to this, and any number may be used.

  The NA 10 is a terminal that accesses information and functions published on the Web. In this embodiment, NA10 assumes portable terminals, such as a smart phone and a tablet, However, It is not limited to this, What is necessary is just an apparatus which can access the internet.

  In the present embodiment, the NA 10 (processing system 1) will be described assuming that the user U1 owns the NA 10 and the user U1 uses the NA 10 for dialogue with the user U2. The NA 10 can be used by three or more users.

  As shown in FIG. 1, the NA 10 includes a voice input unit 11, a GPS (Global Positioning System) receiving unit 13, a communication unit 15, an imaging unit 16, a storage unit 17, an output unit 19, and a control unit 20. Are provided.

  The voice input unit 11 inputs voices uttered by the user U1, user U2, and the like to the NA 10, and can be realized by a sound collector such as a microphone.

  The GPS receiver 13 receives radio waves from GPS satellites and can be realized by a GPS receiver or the like.

  The communication unit 15 communicates with an external device such as the search server 101 via the Internet 107 and can be realized by a communication device such as a NIC (Network Interface Card).

  The imaging unit 16 images the user U1, the user U2, and the like, and can be realized by an imaging device such as a digital camera.

  The storage unit 17 stores various programs executed by the NA 10 and data used for various processes performed by the NA 10. The storage unit 17 is, for example, magnetic, optical, or electrical such as a hard disk drive (HDD), a solid state drive (SSD), a memory card, an optical disk, a read only memory (ROM), and a random access memory (RAM). This can be realized by a storage device that can be stored.

  The output unit 19 outputs the processing result of the control unit 20, and includes a display output display device such as a liquid crystal display and a touch panel display, a sound output sound device such as a speaker, or a print output printing device. For example, these devices may be used in combination.

  The control unit 20 controls each unit of the NA 10, and includes a voice recognition unit 21, a search request unit 25, an acquisition unit 27, a situation recognition unit 29, and an output control unit 33. The voice recognition unit 21, the search request unit 25, the acquisition unit 27, the situation recognition unit 29, and the output control unit 33, for example, cause a processing device such as a CPU (Central Processing Unit) to execute a program, that is, realized by software. Alternatively, it may be realized by hardware such as an IC (Integrated Circuit), or may be realized by using software and hardware together.

  The voice recognition unit 21 recognizes the voice input from the voice input unit 11 and obtains a voice recognition result. FIG. 2 is a block diagram illustrating an example of the configuration of the speech recognition unit 21 according to the present embodiment. As shown in FIG. 2, the speech recognition unit 21 includes an acoustic analysis unit 51, a conversion unit 53, a determination unit 55, and an extraction unit 57.

  The acoustic analysis unit 51 analyzes the voice input from the voice input unit 11 and extracts a feature amount. The conversion unit 53 converts the feature amount extracted by the acoustic analysis unit 51 into text (character string) using dictionary data for speech recognition stored in the storage unit 17. The determination unit 55 determines whether the text converted by the conversion unit 53 is questionable (question sentence) using a natural language processing technique or the like. When the determination unit 55 determines that the extraction unit 57 is questionable, the extraction unit 57 extracts the keyword that is the subject of the question.

  For details of the speech recognition method, for example, a known method disclosed in the above-described Patent Document 1, Japanese Patent Application Laid-Open No. 2004-45591, Japanese Patent Application Laid-Open No. 2008-281901, and the like can be used. Detailed description will be omitted.

  The search request unit 25 requests information search based on the voice recognized by the voice recognition unit 21 (the voice recognition result of the voice recognition unit 21). Specifically, the search request unit 25 requests the search server 101 to search for information when the voice recognition result of the voice recognition unit 21 is questionable.

  For example, it is assumed that the user U1 and the user U2 are talking about history, and the user U1 asks the user U2 “How many years has the Sekigahara battle been?”. Then, it is assumed that the determination unit 55 determines that the system is questionable, and the extraction unit 57 extracts “Sekigahara Battle” and “Year” as the questionable keywords. In this case, the search request unit 25 uses the extracted keywords “Sekigahara Battle” and “Year” as a search query, and requests the search server 101 to search on the web.

  The acquisition unit 27 acquires a search result of information. Specifically, the acquisition unit 27 acquires information search results from the search server 101. For example, when the search server 101 performs a search using “Sekigahara Battle” and “Year” as search queries, the acquisition unit 27 acquires “1600” as a search result and generates a response sentence.

  The situation recognition unit 29 recognizes the situation of the user U1 or the user U2 and obtains a user situation recognition result. Note that the situation recognition unit 29 sequentially recognizes the situation of the user U1 and the user U2. Here, the user situation recognition result includes at least the presence / absence of the user U1 and the user U2, whether or not the user U1 and the user U2 are tilted, and whether or not the user U1 and the user U2 are silent. One of them is applicable.

  The presence / absence of the eyes and whether the head is tilted can be determined by analyzing the image captured by the imaging unit 16. For example, the situation recognizing unit 29 can determine the presence or absence of gaze by analyzing the image captured by the imaging unit 16 and determining whether or not the line of sight of the user U1 or the user U2 is directed to the front. In addition, for example, the situation recognition unit 29 analyzes the image captured by the imaging unit 16 and determines whether the head tilt of the user U1 or the user U2 exceeds a certain angle, thereby tilting the neck. You can determine whether or not.

  Whether or not the user is silent can be determined from the presence or absence of a voice input to the voice input unit 11. For example, the situation recognizing unit 29 monitors the elapsed time after the voice determined to be questionable by the voice recognition unit 21 (determination unit 55) is input to the voice input unit 11, and the next voice is received within a predetermined time. By determining whether or not the input has been made, it can be determined whether or not the user is silent.

  The output control unit 33 is configured to output the search result acquired by the acquisition unit 27 to the output unit 19 based on the user situation recognized by the situation recognition unit 29 (user situation recognition result of the situation recognition unit 29). It is determined whether or not there is a timing to output, and the search result is output to the output unit 19. Note that the output control unit 33 sequentially determines whether it is time to output the search result to the output unit 19 using the user situation recognition result. If the output control unit 33 does not determine that it is time to output the search result to the output unit 19 within a certain period, the output control unit 33 does not cause the output unit 19 to output the search result. In the following description, “timing to be output by the output unit 19” is described as “output timing”.

  For example, the output control unit 33 is generated by the acquisition unit 27 when the user situation recognition result indicates that there is a gaze, indicates that the head is tilted, or indicates that the user is silent. The response section (search result) is output to the output unit 19.

  For example, when the output control unit 33 causes the output unit 19 to perform voice output, the response sentence (search result) generated by the acquisition unit 27 is voice-synthesized and converted into voice, and the output unit 19 outputs the voice. Further, for example, when the output unit 19 performs audio output, the output control unit 33 converts the response sentence (search result) generated by the acquisition unit 27 into drawing data and causes the output unit 19 to output the screen.

  Note that the NA 10 does not have to have all the above-described components as essential components, and may be configured such that some of them are omitted.

  Next, the operation of the processing system of this embodiment will be described.

  FIG. 3 is a flowchart illustrating an example of processing executed by the processing system 1 of the present embodiment.

  First, the voice recognition unit 21 recognizes the voice of the user U1 or the user U2 input from the voice input unit 11 and obtains a voice recognition result (step S101).

  Subsequently, the voice recognition unit 21 determines whether or not the voice recognition result is questionable (step S103). If the voice recognition result is not questionable (No in step S103), the NA 10 does not output a response, and the process returns to step S101.

  When the voice recognition result is questionable (Yes in step S103), the search request unit 25 requests the search server 101 to search on the web with a search query based on the voice recognition result (step S104).

  Subsequently, the search server 101 receives a search query from the NA 10, searches information published on the Web according to the received search query, and transmits the search result to the NA 10 (step S105).

  Subsequently, the acquisition unit 27 acquires a search result of information from the search server 101 (step S107).

  Subsequently, the situation recognition unit 29 sequentially recognizes the situation of the user U1 and the user U2 to obtain a user situation recognition result, and the output control unit 33 uses the user situation recognition result obtained by the situation recognition unit 29. Then, it is determined whether it is the output timing of the search result acquired by the acquisition unit 27 (step S109).

  The situation recognition unit 29 and the output control unit 33 repeat the process of step S109 until the output timing is determined (No in step S109). If the output timing is determined (Yes in step S109), the search result is output. It is made to output to the part 19 (step S111). As a result, it is possible to output at a timing when a response by the NA 10 is desired, and a dialogue promoting effect can be expected.

  If the situation recognition unit 29 and the output control unit 33 do not determine that the search result output timing is within a certain period, the status recognition unit 29 and the output control unit 33 end the processing without causing the output unit 19 to output the search result. As a result, when a response by the NA 10 is not desired, the response is not performed, and it is possible to avoid a hindrance to the dialogue.

  As described above, in the present embodiment, it is determined whether or not it is the output timing using the user situation recognition result, and if it is the output timing, the output is performed, so that a more natural response can be performed by the NA 10.

  In particular, according to the present embodiment, when a question arises when a dialogue is performed between a plurality of users, when another user responds, the NA 10 does not respond and the other users also respond. If not, a response by the NA 10 can be made, so that the interaction between users is not hindered, and a more natural response can be made to the NA 10.

(Modification)
In addition, this invention is not limited to the said embodiment, A various deformation | transformation is possible.

(Modification 1)
In the above embodiment, the search result of the general Web search is used as the response content, but the search result of the movement history of the user U1 or the user U2 may be used as the response content. For example, the location information of the user U1 and the user U2 recognized by the radio wave from the GPS satellite received by the GPS receiving unit 13 is recorded on a cloud providing a storage function on the Web. And when the place visited in the past is asked, the movement history on a cloud may be searched, the information of the said place may be acquired, and you may make it output at an output timing. In this way, it is possible to respond to questions that depend on the user.

(Modification 2)
In the above embodiment, not only the output timing but also the output method may be considered. Specifically, when the output unit 19 can output in a plurality of modes such as display output and audio output, the output control unit 33 determines which mode to output, and in the determined mode public information The narrowing-down result may be output. For example, if the output control unit 33 determines that the location is a public place from the location information of the user U1 and the user U2 recognized by the radio waves from the GPS satellites received by the GPS receiving unit 13, the output control unit 33 does not output sound. It is good also as a display output.

(Hardware configuration)
An example of the hardware configuration of the NA 10 of this embodiment and the modification will be described. The NA 10 according to the present embodiment and the modification includes a control device such as a CPU, a storage device such as a ROM and a RAM, an external storage device such as an HDD, a display device such as a display, a voice input device such as a microphone, and a communication device. And a communication device such as an interface, and has a hardware configuration using a normal computer.

  The programs executed by the NA 10 of this embodiment and the modification are files in an installable format or an executable format, and are CD-ROM, CD-R, memory card, DVD (Digital Versatile Disk), flexible disk (FD). Or the like stored in a computer-readable storage medium.

  Further, the program executed by the NA 10 of the present embodiment and the modification may be provided by being stored on a computer connected to a network such as the Internet and downloaded via the network. Further, the program executed by the NA 10 of this embodiment and the modification may be provided or distributed via a network such as the Internet. Further, the program executed by the NA 10 of this embodiment and the modification may be provided by being incorporated in advance in a ROM or the like.

  The program executed by the NA 10 of this embodiment and the modification has a module configuration for realizing the above-described units on a computer. As actual hardware, the CPU reads out a program from the HDD to the RAM and executes the program, whereby the above-described units are realized on the computer.

1 Processing System 10 NA (Network Agent)
DESCRIPTION OF SYMBOLS 11 Voice input part 13 GPS receiving part 15 Communication part 16 Imaging part 17 Storage part 19 Output part 20 Control part 21 Voice recognition part 25 Search request part 27 Acquisition part 29 Situation recognition part 33 Output control part 51 Acoustical analysis part 53 Conversion part 55 Determination unit 57 Extraction unit 101 Search server 107 Internet

JP 2007-121577 A

Claims (8)

A voice recognition unit that recognizes the user's voice;
A search request unit that requests a search for information based on the voice recognized by the voice recognition unit;
An acquisition unit for acquiring a search result based on a search for information requested by the search request unit;
A situation recognition unit for recognizing the situation of the user;
Based on the situation of the user recognized by the situation recognition unit, it is determined whether it is time to output the search result to the output unit, and when it is time to output, the search result is output to the output unit. An output control unit
A processing apparatus comprising: The situation recognition unit sequentially recognizes the situation of the user,
The processing apparatus according to claim 1, wherein the output control unit sequentially determines whether or not it is a timing to output the search result to the output unit based on a user situation recognized by the situation recognition unit. .   The processing apparatus according to claim 2, wherein the output control unit does not output the search result to the output unit when it is not determined that it is time to output the search result to the output unit within a certain period.   The processing device according to any one of claims 1 to 3, wherein the search request unit requests a search for the information when the voice recognition result is questionable.   The status of the user recognized by the status recognition unit is at least one of the presence / absence of the user's attention, whether or not the user is tilted, and whether or not the user and the conversation partner are silent. The processing apparatus according to any one of claims 1 to 4. A voice recognition unit that recognizes the user's voice;
A search request unit that requests a search for information based on the voice recognized by the voice recognition unit;
A search unit for searching for information requested by the search request unit;
An acquisition unit for acquiring a search result of information of the search unit;
A situation recognition unit for recognizing the situation of the user;
Based on the situation of the user recognized by the situation recognition unit, it is determined whether it is time to output the search result to the output unit, and when it is time to output, the search result is output to the output unit. An output control unit
A processing system comprising: A voice recognition step in which a voice recognition unit recognizes a user's voice;
A search requesting step for requesting a search for information based on the voice recognized by the voice recognition unit;
An obtaining step for obtaining a search result based on a search for information requested by the search requesting unit;
A situation recognition unit that recognizes the situation of the user;
The output control unit determines whether or not it is a timing to output the search result to the output unit based on the situation of the user recognized by the situation recognition unit. An output control step for outputting to the output unit;
Output method including A voice recognition step for recognizing the user's voice;
A search requesting step for requesting a search for information based on the voice recognized by the voice recognition unit;
An acquisition step of acquiring a search result based on a search for information requested by the search request unit;
A situation recognition step for recognizing the situation of the user;
Based on the situation of the user recognized by the situation recognition unit, it is determined whether it is time to output the search result to the output unit, and when it is time to output, the search result is output to the output unit. An output control step to
Program to make the computer function.

Images