JP5705274B2 - Information processing apparatus and method - Google Patents

Description

  The present invention relates to information provision using an information processing apparatus.

  Today, by carrying smart devices (smartphones, tablet PCs, etc.), you can easily search the Internet whenever you need information. An example of facilitating input for search by voice recognition has also been proposed (see, for example, Patent Document 1).

Special table 2003-517158 JP 2009-238199 A

  However, if a search operation is performed whenever information is needed, it has not been possible to provide a topic in real time for the ongoing conversation to promote (that is, play) the conversation.

  Although there is a technique for extracting an important word in conversation and presenting it as text (see, for example, Patent Document 2), text conversion does not provide a topic for conversation although it is recorded. In addition, since the extraction of important words is basically based on the detection of repeated words, there is a large delay from the content of the conversation and not real time. In this respect as well, the conversion of important words into texts did not provide topics for conversation.

  Even if we consider the technology to extract themes from conversations with more complex information processing and provide information, there is the same problem as key word extraction in terms of large delay, and if we predict the development of the discussion to overcome the delay, we can predict It may lead to unforeseen information provision due to the limit of accuracy. Prior to that, complex information processing required large processing power and became an obstacle to implement as a portable device.

  An object of the present invention is to provide a topic for promoting conversation in real time.

  Based on the above object, an information processing apparatus according to one aspect (1) of the present invention includes a voice acquisition unit that acquires a voice of a conversation, an extraction unit that extracts a speech of a conversation partner from the acquired voice, Recognizing means for recognizing words from the extracted utterances, related information acquiring means for acquiring related information based on the speech-recognized words from a server via a communication network, and outputting the acquired related information And an output means.

  An information processing method according to another aspect (6) of the present invention is based on the above-described aspect in the category of the method, and is a voice acquisition process for acquiring a voice of a conversation, and an utterance of a conversation partner from the acquired voice Extraction processing for extracting words, recognition processing for recognizing words from the extracted utterances, related information acquisition processing for acquiring related information based on the speech-recognized words from a server via a communication network, In addition, a computer executes an output process for outputting the related information.

  According to another aspect (2) of the present invention, in any one of the aspects described above, the wearable device includes a microphone, a headphone, and a unit that is attached to a human body.

  According to another aspect (3) of the present invention, in any one of the above aspects, the part of speech determination means for sequentially determining the part of speech of the word that has been speech-recognized by the recognition means; And selecting means for selecting a word as a basis of the related information.

  According to another aspect (4) of the present invention, in any one of the above aspects, the related information acquisition unit acquires the related information based on the speech-recognized word and information representing an environment. Features.

  According to another aspect (5) of the present invention, in any one of the above aspects, the output unit selects an output target by presenting the acquired candidate related information when the utterance speed is less than a predetermined value. When the utterance speed is equal to or higher than a predetermined value, an output target is determined based on a predetermined priority order among the obtained candidates for the related information without obtaining the selection. .

  ADVANTAGE OF THE INVENTION According to this invention, the topic which promotes conversation can be provided in real time.

The functional block diagram which shows a structure about embodiment of this invention. The figure which shows the example of the data in embodiment of this invention. The flowchart which shows the process sequence in embodiment of this invention. The conceptual diagram which shows the example of 1 use in embodiment of this invention.

  Next, a mode for carrying out the present invention (referred to as “embodiment”) will be illustrated along the drawings. In addition, the assumptions common to the contents already described in the background art and problems are omitted as appropriate.

[1. Constitution〕
FIG. 1 shows the configuration of this embodiment. The present embodiment is an information processing apparatus according to the present invention (hereinafter referred to as “a wearable device”) including a microphone M, headphones H (including earphones), and a holder (not shown) that is a means to be worn on a human body. Also referred to as “this device 1”). The search server 2 is a server device that provides information in response to a search request from the outside via a communication network N (Internet, mobile phone network, etc.).

  The apparatus 1 includes a computer configuration, that is, an arithmetic control unit 6 such as a CPU, a storage device 7 such as a main memory and an auxiliary storage device, a communication device 8 (such as a communication device and a communication adapter) with a communication network N, Is provided. The search server 2 also has a computer configuration (not shown), although the specifications are different. In the present apparatus 1, each element shown in FIG. 1 is realized by the arithmetic control unit 6 executing a computer program (not shown) stored in the storage device 7.

  Among the realized elements, the information storage means is not limited to the so-called local storage in the apparatus 1 but may be remote storage by network computing (cloud) or the like. Further, the storage means shown in the present application is a unit that is convenient for explanation and is a main one. The actual storage means may include functions such as input / output and management associated with information storage, the unit of the configuration may be divided or integrated, and other storage means such as a work area may be used as appropriate. Good.

  Of the storage means, the recognition dictionary storage means 35 stores a recognition dictionary for speech recognition (for example, data representing features for each word or sound element). The temporary storage means 45 is means for storing the recognized word together with its part of speech. The search condition storage means 55 stores search conditions for acquiring related information according to words and parts of speech. FIG. 2 shows an example of the temporary storage means 45 and the search condition storage means 55, but the data items are omitted as appropriate. The storage contents of other storage means are not shown.

  The speech synthesis data storage means 66 stores speech data for each word and each sound element for speech synthesis. The part-of-speech dictionary storage unit 75 stores a part-of-speech dictionary for determining the part of speech of a word.

  The arrows in the figure (for example, FIG. 1) exemplify the main direction of the flow of data and control, and do not mean the denial of other flows or the limitation of the direction. Each means other than the storage means is a processing means for realizing or executing an information processing function or action as described below, but these functions or actions are exclusively units for explanation and are implemented in actual hardware. It does not matter if it matches the hardware and software elements.

[2. Action)
FIG. 3 is a flowchart showing the operation of the apparatus 1. FIG. 4 is a conceptual diagram showing an example of utilizing this embodiment.
[2-1. Overview〕
First, an outline of the operation will be described along the flowchart of FIG. The apparatus 1 acquires the voice of the conversation between the user wearing the apparatus 1 as a wearable device ("self" in FIG. 4) and the other party, and the voice acquisition means 20 performs A / D conversion or the like with the microphone M. (Step S11). For example, in the example of FIG. 4, the user's utterance T1 wearing the apparatus 1 and the conversation partner's utterance T2 are both acquisition targets.

  The extraction means 30 extracts the utterance of the conversation partner from the acquired voice based on the difference in input volume based on the difference in distance to the microphone M, the difference in voice frequency, and the like (step S12). In order to extract the speech of the conversation partner, known techniques such as frequency estimation, hidden Markov model, pattern matching, neural network, decision tree, and the like are used. The other party's utterance T2 is extracted, but his own utterance T1 is not extracted.

  The recognition unit 43 recognizes a word from the extracted utterance T2 by comparison with a recognition dictionary in the recognition dictionary storage unit 35 (step S13). For voice recognition, a known technique such as a statistical method, a dynamic time expansion / contraction method, a hidden Markov model, or the like is used. For example, several words such as “this side”, “Italian”, and “ΔΔ” are recognized from the utterance T2 (enclosed by a wavy line in FIG. 4).

  Thereafter, the related information acquisition means 54 acquires related information (for example, glossary or restaurant information) based on the speech-recognized word from the search server 2 using a search API or the like via the communication network (step S18). ). The output unit 65 outputs the acquired related information from the headphones H as synthesized speech using the speech data stored in the speech synthesis data storage unit 66 (step S22).

  In the example of FIG. 4, information of an Italian restaurant in the neighborhood corresponding to the recognized word is output by the synthesized speech V as related information based on the word recognized from the partner's utterance T2. Thanks to this related information, the person wearing this device 1 was able to make a conversation immediately and make a judgment and suggestion of an action (for example, utterance T2). Hereinafter, it will be specifically described including other steps.

[2-2. (Selecting words based on part of speech)
An utterance in a conversation includes many words, but an example of a criterion for selecting a word on which related information is based is the part of speech of the word. That is, the part of speech determination unit 73 sequentially determines the part of speech of the words recognized by the recognition unit 43 using the part of speech dictionary stored in the part of speech dictionary storage unit 75 (step S14). The temporary storage means 45 stores the latest 20 pairs of recognized words and parts of speech determined thereafter (step S15).

  The part of speech is not limited to a part of speech that is generally used (for example, “common noun”, “proprietary noun”, etc.), but also a specialized classification or reserved word for acquiring related information, for example, a reserved word (“this area”, “ It may also be a reserved word (such as “Italian”, “Family”, “Post Office”, “Bank”) or the like (for example, FIG. 2).

  A specific search condition can be associated with a specific part of speech. For example, the related information acquisition unit 54 replaces a reserved word representing a place with information on the current position (for example, “Akasaka, Minato-ku”) measured by GPS or the like in the device 1 and transmits the information to the search server 2 as a search query. To do.

  Among reserved words representing business names, those that are predetermined as representing restaurants (for example, “Italian” and “Family”) have a specific type (for example, the type of search service used for acquiring related information). A search condition of selecting (restaurant search) can also be associated. Information associating the part of speech with the search condition is stored in advance in the search condition storage means 55.

  The selection unit 83 selects a word that is the basis of the related information based on the part of speech of the word stored in the temporary storage unit 45, that is, the part of speech of the word determined in succession (step S17). For example, from the latest 20 sets of words stored in the temporary storage means 45, the one with the highest priority in the order of proper noun> common noun> other is selected.

[2-3. Acquisition of related information based on the environment)
In addition, information related to the environment in which the present apparatus 1 is used (for example, time, place, action content, etc .; also referred to as “environment information”) can be reflected in the acquisition of related information. That is, the related information acquisition unit 54 sequentially updates the environment information (step S16), and acquires the related information based on the word selected by the selection unit 83 and the environment information (step S18).

  For example, even if the program name of a TV program that introduces a restaurant is recognized and selected by voice, the information on nearby restaurants becomes related information in utterances while walking in a restaurant area at lunch time. In the utterance, the TV program information of the program is related information.

[2-4. Proper use of output style)
The output style of related information depends on the speech rate. That is, when the utterance speed is less than the predetermined value (step S19: “NO”), the output unit 65 presents the acquired related information candidates and requests selection of an output target (step S20). The related information candidates include, for example, several tops (three or five) included in the web search result based on a certain word.

  On the other hand, when the utterance speed is equal to or higher than a predetermined value (step S19: “YES”), the output means 65 selects an output target based on a predetermined priority among the acquired related information candidates without asking for selection. Determine (step S21). The predetermined priority order is, for example, the first case in the web search result, the order close to the current position measured by the apparatus 1 in the restaurant search result, the order set by the user in advance, or other order.

[3. effect〕
(1) As described above, in the present embodiment (for example, FIG. 4), the partner can acquire the related information via the network based on the speech-recognized words extracted from the conversation partner, and output the related information. You can provide real-time topics that promote conversation, such as information you don't know about the words you say.

(2) Also, in this embodiment, the information processing apparatus of the present invention is configured as a wearable device (for example, FIG. 1), so that it can be easily used at any time. Can be provided.

(3) Further, in the present embodiment, by selecting words as the basis of the related information from words that are recognized one after another based on the part of speech (for example, steps S14 and S17 in FIG. 3), many words are used as the basis. It is possible to provide relevant information based on appropriate words.

(4) In the present embodiment, TPO (time, place, scene, etc.) is obtained by acquiring related information based on the speech-recognized word and information representing the environment (for example, step S18 in FIG. 3). Appropriate relevant information can be provided.

(5) In this embodiment, if the utterance is not fast (eg, step S19 in FIG. 3: “NO”), the selection of the output target is obtained from the related information candidates (step S20). Step S19: “YES”), it is assumed that the other party is in a hurry or impatient, so by determining the output target with a predetermined priority from the candidates of the related information without seeking selection (Step S21), depending on the situation Provide appropriate information.

[4. Other embodiments]
In addition, the content of the said embodiment and figure is only an illustration, and the presence or absence and arrangement | positioning of each element, the order and content of a process, etc. can be changed suitably. For this reason, this invention also includes the modification illustrated below and other embodiment other than that.

  For example, the information processing apparatus of the present invention is not limited to a wearable device or a device that outputs related information from headphones. The information processing apparatus according to the present invention may be, for example, a smartphone that is put in a pocket or lowered from a neck, an apparatus that outputs related information as visual information from a glasses-type head-mounted display unit, and other forms.

  Moreover, each aspect of the present invention can be understood as other categories (methods, programs, systems including terminals, etc.) that are not specified. In the category of method or program, “means” shown in the category of apparatus is appropriately read as “process” or “step”. In addition, all or any part of “means” can be read as “part” (unit, section, module, etc.).

  Also, the processes and steps shown in the embodiment can be changed by changing the order, executing some of them collectively, or executing them part by part. In addition, hardware elements that implement and execute individual means, processes, and steps may be common, or may differ for each means, process, step, or timing.

  The individual means shown in the present application may be realized by calling a function provided by an external server by an API (Application Program Interface) or network computing (so-called cloud or the like). Furthermore, elements such as means are not limited to computers, and may be realized by other information processing mechanisms that appear now or in the future.

1. Information processing device (this device)
2 Search server 6 Arithmetic control unit 7 Storage device 8 Communication device 20 Voice acquisition means 30 Extraction means 35 Recognition dictionary storage means 43 Recognition means 45 Temporary storage means 54 Related information acquisition means 55 Search condition storage means 65 Output means 66 Speech synthesis data storage Means 73 Part of speech judgment means 75 Part of speech dictionary storage means 83 Selection means H Headphone M Microphone N Communication network T1, T2, T3 Utterance V Related information

Claims (5)

Voice acquisition means for acquiring voice of conversation;
Extracting means for extracting the utterance of the conversation partner from the acquired voice;
Recognizing means for recognizing words from the extracted utterances;
Related information acquisition means for acquiring related information based on the speech-recognized word from a server via a communication network;
Output means for outputting the acquired related information;
Equipped with a,
The output means presents a candidate for the related information obtained when the speed of the utterance is less than a predetermined value and asks for selection of an output target, and asks for the selection when the speed of the utterance exceeds a predetermined value And determining an output target based on a predetermined priority order among the acquired candidates for the related information . A microphone,
With headphones,
Means for wearing on the human body;
The information processing apparatus according to claim 1, wherein the information processing apparatus is a wearable device. Part-of-speech determination means for sequentially determining the part-of-speech of the word speech-recognized by the recognition means;
Selection means for selecting a word that is the basis of the related information based on the part of speech of the words that are determined in succession;
The information processing apparatus according to claim 1, further comprising:   4. The information processing according to claim 1, wherein the related information acquisition unit acquires the related information based on the speech-recognized word and information representing an environment. 5. apparatus. Voice acquisition processing to acquire the voice of the conversation;
An extraction process for extracting the utterance of the conversation partner from the acquired voice;
Recognition processing for recognizing words from the extracted utterances;
Related information acquisition processing for acquiring related information based on the speech-recognized word from a server via a communication network;
An output process for outputting the acquired related information;
The computer runs,
In the output process, when the utterance speed is less than a predetermined value, the obtained candidate of the related information is presented to request selection of an output target, and when the utterance speed is equal to or higher than the predetermined value, the selection is requested. And determining an output target based on a predetermined priority order among the acquired candidates for the related information .

Images