JP2017009685A - Information processing device, information processing method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To appropriately perform a translation.SOLUTION: An information processing device includes an acquisition part for acquiring first sentence information indicating a sentence of a first language, a translation part for executing translation processing for generating second sentence information indicating a sentence obtained by translating the sentence indicated by the first sentence information into a second language different from the first language, a second evaluation part for evaluating the existence/absence of abnormality in the translation processing, and an operation control part for causing a character to operate on the basis of a second evaluation by the second evaluation part. Also, the information processing device includes, for example a first evaluation part for evaluating the existence/absence of abnormality on voice recognition processing for generating the first sentence information from voice information indicating a voice of the first language, the acquisition part acquires the first sentence information by executing the voice recognition processing, and the operation control part causes the character to operate on the basis of either or both of the first evaluation by the first evaluation part and the second evaluation by the second evaluation part.SELECTED DRAWING: Figure 1

Description

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

Conventionally, an apparatus for translating a sentence in one language into a sentence in another language has been developed. In order to improve the convenience of such machine translation, it is considered to improve translation accuracy, evaluate the quality of translation, and the like.
For example, in Patent Document 1, a data conversion device in which data conversion means from first data to second data and data reverse conversion means from second data to first data coexist is used. On the other hand, a data conversion suitability evaluation method for evaluating a conversion suitability in a data conversion means and calculating a conversion suitability value, wherein the data conversion means converts the first data and obtains the converted second data A conversion step, a data reverse conversion step of reversely converting the converted second data by the data reverse conversion means to obtain the first data after reverse conversion, and the similarity between the first data and the post-inverse converted first data A similarity calculation step for inputting the calculation means to calculate the similarity by a predetermined similarity calculation formula, and converting the output from the output means as the conversion suitability value in the data conversion means for the first data. Data conversion qualification method characterized by comprising the value output steps are disclosed.

JP 2006-252323 A

  By the way, when providing a translation service such as an interpreter by applying a speech recognition technology or a speech synthesis technology, there are various causes for the difficulty of translation. For example, since the utterance is unclear, it may be difficult to specify the content of the utterance, or even if the utterance content can be specified, it may be difficult to generate a translation. In such a case, by explaining the cause of the difficulty of translation to the speaker and how to deal with it, prompting the user to input speech that is easier to translate and re-speaking the speech, in line with the intention of the speaker Translation can be done properly. However, the reason why translation is difficult and how to deal with it have not been shown to the user in an easy-to-understand manner. Therefore, there are cases where the translation cannot be performed properly.

  An object of some aspects of the present invention is to provide an information processing apparatus, an information processing method, and a program that can appropriately perform translation.

  Another object of another aspect of the present invention is to provide an information processing apparatus, an information processing method, and a program that can achieve the effects described in the embodiments described later.

  (1) In order to solve the above-described problem, according to one aspect of the present invention, an acquisition unit that acquires first sentence information indicating a sentence in a first language, and a sentence indicated by the first sentence information are included in the first sentence. A translation unit for executing a translation process for generating second sentence information indicating a sentence translated into a second language different from the language; a second evaluation unit for evaluating presence or absence of abnormality in the translation process; and the second evaluation unit And an action control unit that moves the character based on the second evaluation.

  (2) Moreover, 1 aspect of this invention is provided with the 1st evaluation part which evaluates the presence or absence of abnormality in the speech recognition process which produces | generates the said 1st sentence information from the audio | voice information which shows the audio | voice of the said 1st language, The acquisition unit acquires the first sentence information by executing the voice recognition process, and the operation control unit performs a first evaluation by the first evaluation unit and a second evaluation by the second evaluation unit. Is an information processing apparatus that moves a character based on one or both of the above.

  (3) Moreover, one aspect of the present invention is the information processing apparatus according to (2), in which the operation control unit has no abnormality in the speech recognition process and no abnormality in the translation process In addition, the character is operated based on the second sentence information.

  (4) Moreover, one aspect of the present invention is the information processing apparatus according to (2) or (3), in which the operation control unit has no abnormality in the speech recognition processing and the translation processing. When there is an abnormality, the character is moved based on the first sentence information.

  (5) Moreover, 1 aspect of this invention is an information processing apparatus in any one of (2) to (4), Comprising: Both the said 1st evaluation and the said 2nd evaluation, or either, The said character And a motion information storage unit that stores different motion specification information for each language, and the motion control unit is provided in the first language or the second language. By referring to the corresponding action definition information, an action corresponding to one or both of the first evaluation and the second evaluation is selected, and the selected action is performed by the character.

  (6) Moreover, one aspect of the present invention is the information processing apparatus according to (5), wherein the operation definition information is different for each use scene and is not defined in the operation definition information for each language. A scene action storage unit that stores action definition information that defines a correspondence relationship is provided, and the action control unit refers to the action definition information according to the usage scene of the device itself, and the first evaluation or the second evaluation is performed. Or an action corresponding to both of them is selected, and the character is caused to perform the selected action.

  (7) One embodiment of the present invention is the information processing apparatus according to (6), in which an operation reception unit that receives an operation for setting the operation definition information and the operation that the operation reception unit receives And a motion registration unit that stores the motion regulation information in one or both of the language motion information storage unit and the scene motion storage unit.

(8) According to one aspect of the present invention, the information processing device acquires first sentence information indicating a sentence in the first language, and the information processing apparatus indicates the sentence indicated by the first sentence information. A second step of executing a translation process for generating second sentence information indicating a sentence translated into a second language different from the first language, and the information processing apparatus evaluates whether there is an abnormality in the translation process And a fourth step in which the information processing apparatus moves the character based on the evaluation in the third step.
It is.

  (9) Further, according to one aspect of the present invention, a first step of acquiring, in a computer, first sentence information indicating a sentence in a first language, a sentence indicated by the first sentence information, Based on the second step of executing a translation process for generating second sentence information indicating a sentence translated into a different second language, the third step of evaluating the presence or absence of abnormality in the translation process, and the evaluation in the third step And a fourth step of moving the character.

  According to the embodiment of the present invention, translation can be appropriately performed.

It is a figure for demonstrating the outline | summary of the information processing apparatus which concerns on the 1st Embodiment of this invention. It is a figure showing an example of operation regulation information concerning the embodiment. It is a figure showing an example of a character's operation concerning the embodiment. It is a figure which shows the outline | summary of the movement pattern of the character which concerns on the embodiment. It is a figure which shows an example of the hardware constitutions of the information processing apparatus which concerns on the embodiment. It is a figure which shows an example of a function structure of the information processing apparatus which concerns on the embodiment. It is a flowchart which shows an example of the flow of a process by the information processing apparatus which concerns on the embodiment. It is a figure for demonstrating the outline | summary of the information processing apparatus which concerns on the 2nd Embodiment of this invention. It is a figure which shows an example of a function structure of the information processing apparatus which concerns on the embodiment. It is a figure for demonstrating the outline | summary of the information processing apparatus which concerns on the 3rd Embodiment of this invention. It is a figure which shows an example of a function structure of the information processing apparatus which concerns on the embodiment.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
[First embodiment]
[Outline of information processing equipment]
A first embodiment of the present invention will be described. First, an overview of the information processing apparatus 10 according to the present embodiment will be described.
The information processing apparatus 10 (FIG. 6) according to the present embodiment is an apparatus that supports interpretation of conversations in two or more languages different from each other. That is, there are a plurality of users in the information processing apparatus 10, and these plurality of users each have a conversation in different languages. However, when there are three or more users, the number of languages used by these three or more users may be two or more. For example, if there are three users, two of them may use the same language.

FIG. 1 is a diagram for explaining an overview of the information processing apparatus 10 according to the present embodiment.
FIG. 1 shows three scenes C11 to C13 related to the use of the information processing apparatus 10. In these first to third scenes C11 to C13, voice reception units 131-1 and 131-2, a character CR, a speaker SP, and a receiver RE are shown, respectively.
The speaker SP is a user on the utterance side in a conversation by a plurality of users among the plurality of users of the information processing apparatus 10.
The listener RE is a user on the receiver side in a conversation between the plurality of users among the plurality of users of the information processing apparatus 10. Although FIG. 1 shows an example in which there is one speaker SP and one listener RE, respectively, the number of users of the information processing apparatus 10 may be three or more. However, even in this case, there is basically one speaker SP. The relationship between the speaker SP and the receiver RE is not fixed, and the speaker SP and the receiver RE may be switched according to the progress of the conversation. In the present embodiment, as an example, a case will be described in which the language used by the speaker SP for conversation is Japanese, and the language used by the receiver RE for conversation is English.

  The character CR is a character that plays the role of an interpreter in a conversation between a plurality of users of the information processing apparatus 10. The movement of the character CR is controlled by the information processing apparatus 10. The character CR may be a real object such as a robot, or may be a character animation displayed on a screen, for example. When displaying the character CR on the screen, the character CR may be drawn two-dimensionally or three-dimensionally. The character CR may have any shape as long as it can express movement. Hereinafter, as an example, a case where the character CR is an image of a humanoid robot displayed on the display unit included in the information processing apparatus 10 will be described.

  The voice receiving units 131-1 and 131-2 are devices such as microphones that collect conversations by a plurality of users of the information processing apparatus 10. Although FIG. 1 shows two voice reception units 131-1 and 131-2 installed for each user, the voice reception units 131-1 and 131-2 may not be provided for each user. Hereinafter, when it is not necessary to particularly distinguish the voice receiving units 131-1 and 131-2, they are collectively referred to as the voice receiving unit 131.

Next, the first scene C11 to the third scene C13 shown in FIG. 1 will be described. In the example shown in FIG. 1, each scene progresses in time series in the order of a first scene C11, a second scene C12, and a third scene C13.
In the first scene C11, the speaker SP is speaking in Japanese. This speech is collected by the voice reception unit 131-1. Here, the information processing apparatus 10 analyzes the content spoken by the speaker SP from the collected sound. And the 1st sentence information showing the 1st sentence (text) of the analyzed utterance content is produced | generated. That is, the first sentence is a sentence representing the content uttered by the speaker SP in the language used by the speaker SP. In addition, the information processing apparatus 10 generates second sentence information representing a second sentence (text) obtained by translating the first Japanese sentence into English. That is, the second sentence is a sentence representing the content uttered by the speaker SP in the language used by the receiver RE. The information processing apparatus 10 evaluates the presence or absence of abnormality in these speech recognition processes and translation processes.

  Here, the evaluation of the presence / absence of abnormality in the speech recognition processing is the certainty in writing the content of speech from speech information. Further, the evaluation of the presence or absence of abnormality in the speech recognition process is also an evaluation of the accuracy of the first sentence. The accuracy of the first sentence is the degree of matching between the statement content and the first sentence. For example, when it is considered that the content of the statement is well matched with the first sentence, the accuracy of the first sentence is high, and it is highly likely that the content of the comment is not consistent with the first sentence, or it is difficult to generate the first sentence information. In this case, the accuracy of the first sentence is low. The information processing apparatus 10 determines that there is an abnormality in the speech recognition process when the accuracy of the first sentence is lower than a predetermined threshold. Hereinafter, the evaluation of the presence or absence of abnormality in the voice recognition process may be referred to as a voice recognition evaluation process.

  Moreover, it can be said that the evaluation of the presence or absence of abnormality in the translation process is the certainty of translation from the first sentence to the second sentence. Also, the evaluation of the presence or absence of abnormality in the translation process is to evaluate the accuracy of the second sentence. The accuracy of the second sentence is the accuracy of translation of the second sentence. For example, when the meanings of the first sentence and the second sentence are considered to match in many parts, the accuracy of the second sentence information is high, and the meanings of the first sentence and the second sentence may not match very much. The accuracy of the second sentence information is low when there is a possibility that there is a plurality of options in the translation from the first sentence to the second sentence and there is a possibility that the translation is erroneous. When the accuracy of the second sentence is lower than the predetermined threshold, the information processing apparatus 10 determines that there is an abnormality in the translation process. Hereinafter, the evaluation of the presence or absence of abnormality in the translation process may be referred to as a translation evaluation process.

  As an evaluation method of the accuracy of the first sentence and the accuracy of the second sentence, any already disclosed method may be used. For example, the certainty factor can be used as an evaluation index of the accuracy of the first sentence and the second sentence. The certainty factor in the speech recognition evaluation process is an index representing the likelihood of the first sentence generated from the speech information. For example, the information processing apparatus 10 compares the voice information received by the voice receiving unit 131 with the voice information of a template prepared in advance, thereby specifying the content of the statement. At this time, the information processing apparatus 10 can calculate the certainty factor by calculating the degree of coincidence between the audio information received by the audio receiving unit 131 and the audio information of the template.

  In addition, the certainty factor in the translation evaluation process is an index representing the probability of the second sentence generated from the first sentence. For example, the information processing apparatus 10 back-translates the second sentence from the second language to the first language. Then, the information processing apparatus 10 can acquire the certainty level by comparing the sentence in the first language of the translated word with the original first sentence and calculating the degree of coincidence between the two sentences. Hereinafter, the evaluation by the voice recognition evaluation process may be referred to as a first evaluation. Moreover, the evaluation by the translation evaluation process may be referred to as a second evaluation.

  Next, in the second scene C12, the character CR reacts to the speech of the speaker SP in the first scene C11. Here, the information processing apparatus 10 selects the action of the character CR according to the first evaluation or the second evaluation. For example, in the first scene C11 described above, since the voice of the speaker SP is low, the information processing apparatus 10 determines that the accuracy of the first sentence is low. Accordingly, in the second scene C12, for example, the information processing apparatus 10 causes the character CR to perform an action of spreading the hand in front of the mouth to the speaker SP, and prompts the speaker to speak with a loud voice. Thereby, since the speaker SP can recognize that the voice is low, it is possible to take a countermeasure such as re-speaking with a louder voice.

  Next, in the third scene C13, the information processing apparatus 10 can recognize the utterance content by re-phrasing the speaker SP in the second scene C12, and thus generates a sentence in which the content is translated into English. Then, the information processing apparatus 10 outputs the English translation as a voice while making the character CR perform a gesture according to the sentence. Thereby, the listener RE can recognize the content spoken by the speaker SP in Japanese in correct English.

As described above, the information processing apparatus 10 according to the present embodiment generates the first sentence information indicating the first language sentence from the voice information indicating the first language (for example, Japanese) voice. Execute. In addition, the information processing apparatus 10 executes a translation process for generating second sentence information indicating a sentence obtained by translating a sentence indicated by the first sentence information into a second language (for example, English) different from the first language. In addition, the information processing apparatus 10 evaluates whether there is an abnormality in the speech recognition process and the translation process. Then, the information processing apparatus 10 moves the character CR based on the evaluation result.
As a result, the information processing apparatus 10 can cause the character CR to perform an appropriate action, and can easily indicate to the user the cause that makes translation difficult and how to deal with it. Therefore, the information processing apparatus 10 can appropriately perform translation.

[Operation regulation information]
Next, the operation definition information processed by the information processing apparatus 10 according to the present embodiment will be described.
The action definition information is information that defines an action according to the evaluation result of the speech recognition process or the translation process.
First, the data structure of the operation defining information will be described.
FIG. 2 is a diagram illustrating an example of the operation definition information according to the present embodiment.
The action defining information is information that defines the action of the character CR according to the conversation situation. FIG. 2 shows, as an example, action definition information when the character CR is moved according to the first evaluation and the second evaluation. In the example shown in FIG. 2, the action definition information includes classification information (“classification” in FIG. 2), physical input information (“physical input” in FIG. 2), and voice identification information (“voice identification” in FIG. 2). Speech recognition semantic analysis information ("voice recognition semantic analysis" in FIG. 2), text analysis information ("text analysis" in FIG. 2), grammar analysis information ("grammar analysis" in FIG. 2), and target language Decoding information (“target language decoding” in FIG. 2), completion determination information (“completion determination information” in FIG. 2), action ID (IDentifier) (“action ID” in FIG. 2), and assumed situation information (FIG. 2 and the “expected situation” in FIG.

The classification information is information representing the classification between the first evaluation and the second evaluation. In other words, the classification information is information for identifying each record of the operation definition information.
The physical input information is information indicating whether or not there is an abnormality in the physical characteristics of the collected voice in the generation of the first sentence information. In the physical input information, when there is an abnormality in the quality of the collected voice, details of the abnormality are described. The determination of whether or not there is an abnormality in the sound quality is performed by, for example, measuring the sound quality and comparing the measured value with a predetermined threshold value. Thus, physical input information is an example of information representing the first evaluation. That is, in this embodiment, the absence of abnormality does not necessarily mean that there is no abnormality at all. For example, it may be determined that there is no abnormality when there is a slight abnormality that does not interfere with the processing, or when the subsequent processing is partially possible.

  The voice identification information is information indicating whether or not there is an abnormality in the clarity of the collected voice in the generation of the first sentence information. In the voice identification information, when there is an abnormality in the clarity of the collected voice, details of the abnormality are described. The determination of the presence or absence of an abnormality in speech clarity is performed, for example, by measuring the clarity of speech and comparing the measured value with a predetermined threshold value. Thus, the voice identification information is an example of information representing the first evaluation.

  The speech recognition semantic analysis information is information indicating the presence or absence of abnormality in the linguistic expression of the first sentence generated from the speech. In the speech recognition semantic analysis information, when there is an abnormality in the linguistic expression of the first sentence, details of the abnormality are described. The determination of the presence or absence of abnormality in the linguistic expression of the first sentence is performed, for example, by measuring the linguistic expression of the first sentence and comparing the measured value with a predetermined threshold value. As described above, the speech recognition semantic analysis information is an example of information representing the first evaluation.

The text analysis information is information indicating whether or not there is an abnormality in text analysis related to the translation of the first sentence in the generation of the second sentence information. The determination of the presence / absence of text analysis abnormality is performed, for example, by measuring the likelihood of the first sentence and comparing the measured value with a predetermined threshold. As described above, the text analysis information is an example of information indicating the second evaluation.
The grammatical analysis information is information indicating the presence or absence of abnormality in the grammatical analysis of the first sentence in the generation of the second sentence information. The determination of the presence or absence of abnormality in the grammatical analysis of the first sentence is performed, for example, by measuring the grammatical correctness of the first sentence and comparing the measured value with a predetermined threshold value. As described above, the grammatical analysis information is an example of information indicating the second evaluation.

The target language decoding information is information indicating the presence or absence of an abnormality in translation conversion from the first sentence to the second sentence in the generation of the second sentence information. In the target language decoding information, when there is an abnormality in translation conversion, details of the abnormality are described. Thus, the target language decoding information is an example of information representing the second evaluation.
The completion determination information is an example of information indicating the presence / absence of abnormality in the voice output stage of the second sentence.

The action ID is information for identifying the action of the character CR.
Assumed information refers to the state of conversation and translation assumed based on the values of physical input information, speech identification information, speech recognition semantic analysis information, text analysis information, grammar analysis information, target language decoding information, and completion determination information. It is information to represent.
As will be described later, the information processing apparatus 10 may operate the character CR based on the first sentence or the second sentence. In this case, for example, an operation corresponding to an expression such as a word, a phrase, or a clause is described in the operation definition information.

FIG. 3 is a diagram illustrating an example of the motion of the character CR according to the present embodiment.
FIG. 3 shows a correspondence relationship between the operation (action) and the meaning of the operation for each action ID described above. The information processing apparatus 10 specifies an action ID corresponding to the first evaluation and the second evaluation based on the action defining information, and causes the character CR to perform an action corresponding to the action ID.

  As described above, each record of the operation definition information defines an operation according to the state of speech assumed from the first evaluation and the second evaluation and the processing abnormality. Therefore, the information processing apparatus 10 can cope with a situation where translation is difficult by moving the character CR based on the action defining information.

FIG. 4 is a diagram showing an outline of the motion pattern of the character CR according to this embodiment.
Here, the operation pattern is a classification of a plurality of operations based on specific criteria.
In general, there is a correlation between language and cultural sphere. In general, gestures tend to be similar for each cultural area. Accordingly, the character CR can be appropriately operated by determining the movement pattern of the character CR for each language used for translation.

  In addition, the gesture may differ depending on the situation where the conversation is performed in addition to the cultural area. That is, the gesture may differ depending on the scene where the information processing apparatus 10 is used. Specifically, the gestures used between friends are different from the gestures used between employees and customers in the store. Accordingly, the character CR can be appropriately operated by determining the operation pattern of the character CR for each use scene (service, situation).

As described above, the action regulation information is prepared for each language and each usage scene, and the action CR is appropriately operated by selecting the action regulation information according to the language used for the conversation or the scene where the information processing apparatus 10 is used. Can be made.
In addition, since the difference in gestures largely depends on the cultural sphere, the action definition information may be prepared for each language, and the action definition information for each use scene may be prepared as a difference file. That is, as shown in FIG. 4, the operation pattern may be defined hierarchically. Specifically, an operation pattern for each language may be used as a basis, and a characteristic operation pattern for each usage scene, that is, an operation pattern unique to a service may be defined as an exception or application. Thereby, the motion pattern of the character CR can be efficiently defined. In the definition of such an operation pattern, the hierarchy may be a plurality of three or more. Further, the classification of each layer may be arbitrarily determined in addition to language and service.

[Configuration of information processing device]
Next, the configuration of the information processing apparatus 10 will be described.
First, the hardware configuration of the information processing apparatus 10 will be described.
FIG. 5 is a diagram illustrating an example of a hardware configuration of the information processing apparatus 10 according to the present embodiment.
The information processing device 10 is an electronic device such as a personal computer, a mobile phone, a tablet, a smartphone, a PHS (Personal Handy-phone System) terminal device, or a PDA (Personal Digital Assistant).

The information processing apparatus 10 includes a CPU (Central Processing Unit) 11, a storage unit 12, an input unit 13, an audio output unit 14, and a display unit 15.
The storage unit 12 includes, for example, an HDD (Hard Disk Drive), an SSD (Solid State Drive), an EEPROM (Electrically Erasable Programmable Read-Only Memory), and a ROM (Read-Only Memory). A program executed by the CPU 11, various information processed by the CPU 11, processing results by the CPU 11, and the like are stored.

The input unit 13 includes various input devices such as a pointing device such as a mouse and a touch panel, a keyboard, and a microphone, for example. The content of the operation that can be received by the input unit 13 may be displayed by, for example, a display device included in the display unit 15.
The audio output unit 14 includes, for example, an audio device such as a speaker.
The display unit 15 includes a display device such as a liquid crystal display or an organic EL (Electro-Luminescence) display.
The above is the description of the hardware configuration of the information processing apparatus 10.

Next, the functional configuration of the information processing apparatus 10 will be described.
FIG. 6 is a diagram illustrating an example of a functional configuration of the information processing apparatus 10 according to the present embodiment.
The information processing apparatus 10 includes a storage unit 12, a voice output unit 14, a display unit 15, a control unit 110, a voice reception unit 131, and an operation reception unit 132.
The storage unit 12 includes an operation information storage unit 121 that stores operation definition information. The motion information storage unit 121 includes a language motion information storage unit 122 and a scene motion information storage unit 123.

The language action information storage unit 122 stores action definition information for each language. The operation definition information for each language defines, for example, all basic operation patterns common to cultural areas corresponding to various languages.
The scene motion information storage unit 123 stores motion regulation information for each usage scene. For example, among the operation patterns common to services corresponding to various usage scenes, the operation definition information for each use scene defines a service-specific operation pattern that is not described in the above-described operation definition information for each language. For example, the action definition information for each use scene may be prepared as a differential file separated from the action definition information for each language.

The voice reception unit 131 receives a voice input. Specifically, for example, the voice reception unit 131 collects voice around the information processing apparatus 10. That is, the voice reception unit 131 collects the speech of the speaker SP. The voice reception unit 131 outputs voice information indicating the collected voice to the control unit 110.
The operation receiving unit 132 receives an operation input. Specifically, for example, the operation accepting unit 132 accepts selection of a language used by the user of the information processing apparatus 10, a scene where the information processing apparatus 10 is used, and the like. Moreover, the operation reception part 132 may receive the input of the action regulation information which a user uniquely defines. That is, the operation accepting unit 132 may accept an operation for setting the action defining information. The operation reception unit 132 outputs operation information indicating the received operation to the control unit 110.

The control unit 110 controls operations of various configurations included in the information processing apparatus 10. The control unit 110 is realized, for example, when the CPU 11 executes a program stored in the storage unit 12.
The control unit 110 includes a speech recognition unit (acquisition unit) 111, a translation unit 112, a speech synthesis unit 113, a first evaluation unit 114, a second evaluation unit 115, a motion control unit 116, and a motion registration unit 117. And comprising.

  The voice recognition unit 111 executes voice recognition processing. Specifically, the voice recognition unit 111 acquires voice information from the voice reception unit 131. The voice recognition unit 111 analyzes the acquired voice information and converts the speech of the speaker SP included in the voice information into text. That is, the voice recognition unit 111 generates first sentence information indicating the first sentence. The voice recognition unit 111 outputs the generated first sentence information to the translation unit 112.

  The translation unit 112 performs a translation process. Specifically, the translation unit 112 acquires first sentence information from the voice reception unit 131. The translation unit 112 analyzes the acquired first sentence information and translates the sentence indicated by the first sentence information into the language of the listener RE. That is, the translation unit 112 generates second sentence information indicating the second sentence. The translation unit 112 outputs the generated second sentence information to the speech synthesis unit 113.

The speech synthesizer 113 outputs the translation result. Specifically, the speech synthesizer 113 acquires the second sentence information from the translator 112. The voice synthesizer 113 generates voice information obtained by converting the second sentence indicated by the acquired second sentence information into a voice. The voice synthesizer 113 outputs the generated voice information to the voice output unit 14. Thereby, the voice of the first language input to the voice receiving unit 131 is uttered from the voice output unit 14 as the voice of the second language different from the first language.
In addition, when an abnormality is found in the first evaluation and the second evaluation, the voice synthesizer 113 outputs the content of the abnormality in voice according to the action of the character CR.

  The first evaluation unit 114 performs a speech recognition evaluation process. Specifically, the first evaluation unit 114 generates first evaluation information such as the above-described physical input information, voice identification information, and voice recognition semantic analysis information in each process of the voice recognition processing. The first evaluation unit 114 outputs the generated first evaluation information to the operation control unit 116.

  The second evaluation unit 115 executes a translation evaluation process. Specifically, the second evaluation unit 115 generates second evaluation information such as the text analysis information, the grammar analysis information, and the target language decoding information described above in each process of the translation process. The second evaluation unit 115 outputs the generated second evaluation information to the operation control unit 116.

The motion control unit 116 controls the motion of the character CR. The operation control unit 116 mainly selects operation defining information and specifies an operation.
First, selection of the operation regulation information will be described. Based on the operation information acquired from the operation receiving unit 132, the motion control unit 116 selects the motion regulation information used for determining the motion of the character CR. This operation information includes, for example, information on the language used by the speaker SP selected by the user of the information processing apparatus 10, information on the language used by the receiver RE, and information on the usage scene of the information processing apparatus 10. Yes. The operation control unit 116 acquires the language selected by the user and the operation specification information of the use scene from the operation specification information for each language and use scene stored in the operation information storage unit 121. Accordingly, the motion control unit 116 can cause the character CR to perform a motion suitable for each language zone and usage scene.

  Note that the language of the speaker SP may be specified in the voice recognition processing by the voice recognition unit 111. Even in this case, the motion control unit 116 can select the motion regulation information according to the language of the speaker SP, so that the user of the information processing apparatus 10 can perform an unnecessary operation without performing the unnecessary operation. It is possible to cause the character CR to perform an appropriate action according to the SP language.

  Next, the operation specification will be described. The operation control unit 116 acquires first evaluation information from the first evaluation unit 114. In addition, the operation control unit 116 acquires second evaluation information from the second evaluation unit 115. The operation control unit 116 refers to the operation defining information selected by the above-described processing, and specifies the action ID corresponding to the value of the first evaluation information and the value of the second evaluation information. That is, the operation control unit 116 specifies an operation according to the first evaluation and the second evaluation. At this time, when the character control unit 116 causes the character CR to move with respect to the speaker SP, the operation control unit 116 specifies the operation with reference to the action defining information according to the language of the speaker SP. In addition, when the character CR is operated with respect to the listener RE, the motion is specified with reference to the action definition information according to the language of the receiver RE. Then, the motion control unit 116 causes the character CR to perform the identified motion. This operation is coordinated with, for example, audio output by the audio output unit 14. As a result, the information processing apparatus 10 can further indicate to the user the cause of the difficulty of translation and the coping method.

In addition, the operation control unit 116 may select an operation according to the contents by referring to the first sentence information and the second sentence information. For example, the motion control unit 116 displays an image of a noun (for example, cooking) that appears in the first sentence or the second sentence, or causes the character CR to perform an action that expresses emotion according to the exclamation. Also good. Thereby, since the information processing apparatus 10 can operate the character CR according to the utterance content of the speaker SP, the conversation can be more smoothly advanced.
The selection of the operation based on the first sentence information may be performed, for example, when an abnormality is detected in the translation process. Further, even when an abnormality is detected in the speech recognition process or the translation process, the character CR may be operated based on the portion that has been converted into text. Thereby, the information processing apparatus 10 can notify the receiver RE of at least a part of the speech content of the speaker SP even when the speech content of the speaker SP cannot be translated.

  Further, the motion control unit 116 may operate the character CR before the voice recognition processing and the translation processing are all completed. For example, in the voice recognition process, the character CR may be operated based on a part of the voice information acquired from the voice reception unit 131 where the voice recognition is completed (that is, a part that can be heard).

  Further, the motion control unit 116 may control the direction of the character CR. In this case, when the user of the information processing apparatus 10 speaks, the motion control unit 116 controls the direction of the character CR so that the character CR faces the user. In addition, the motion control unit 116 controls the character CR to face the listener RE when the speech is finished. The arrangement of each of the plurality of users of the information processing apparatus 10 may be specified by an arbitrary method. The arrangement of each user may be specified based on, for example, the strength of the voice input received by the voice receiving unit 131, or may be specified by image recognition from the captured image of the imaging unit included in the information processing apparatus 10. .

  Further, the number of character CRs may not be one for each information processing apparatus 10. For example, the information processing apparatus 10 may control the character CR corresponding to each of a plurality of users. Further, the information processing apparatus 10 may control the character CR corresponding to each of the first evaluation and the second evaluation. Specifically, for example, a character CR that operates when an abnormality in the speech recognition process is detected and a character CR that operates when an abnormality in the translation process is detected may be prepared. As a result, the information processing apparatus 10 can further indicate to the user the cause of the difficulty of translation and the coping method.

The operation registration unit 117 generates operation definition information based on the operation received by the operation reception unit 132. At this time, for example, the operation registration unit 117 displays the type of operation and the type of abnormality that can be detected in the speech recognition process and the translation process in a selectable manner on the display unit 15, and the action definition information based on the selection by the user Is generated. Then, the operation registration unit 117 stores the generated operation definition information in the operation information storage unit 121. Note that the operation registration unit 117 may be able to register the operation definition information corresponding to the language, the usage scene, or the like, or may be able to register the operation definition information corresponding to each user. Further, the motion information storage unit 121 may edit the motion regulation information already stored in the motion information storage unit 121.
The above is the description of the configuration of the information processing apparatus 10.

[Operation of information processing device]
Next, the operation of the information processing apparatus 10 will be described.
FIG. 7 is a flowchart showing an example of the flow of processing by the information processing apparatus 10 according to the present embodiment.
Here, as an example, it is a flowchart showing the flow of processing until the speaker SP utters and causes the character CR to perform an action corresponding to the utterance.
(Step S100) The voice receiving unit 131 receives a voice input by the speaker SP. The voice reception unit 131 outputs voice information indicating the received voice to the voice recognition unit 111. Thereafter, the control unit 110 proceeds with the process to step S102.

(Step S <b> 102) The voice recognition unit 111 acquires voice information from the voice reception unit 131. The voice reception unit 131 analyzes the acquired voice information and generates first sentence information. The voice recognition unit 111 outputs the generated first sentence information to the translation unit 112. Thereafter, the control unit 110 advances the process to step S104.
(Step S104) The first evaluation unit 114 evaluates whether or not there is an abnormality in the speech recognition process, and generates first evaluation information. The voice recognition unit 111 outputs the generated first evaluation information to the operation control unit 116. Thereafter, control unit 110 advances the process to step S106.

(Step S <b> 106) The translation unit 112 acquires first sentence information from the speech recognition unit 111. The translation unit 112 analyzes the acquired first sentence information and generates second sentence information indicating the second sentence. Thereafter, control unit 110 advances the process to step S108.
(Step S108) The second evaluation unit 115 evaluates the presence or absence of abnormality in the translation process, and generates second evaluation information. The translation unit 112 outputs the generated second evaluation information to the operation control unit 116. Thereafter, the control unit 110 advances the process to step S110.

(Step S110) The operation control unit 116 refers to the operation definition information stored in the operation information storage unit 121, and selects an operation based on the first evaluation information and the second evaluation information. At this time, the operation control unit 116 may refer to the first sentence information and the second sentence information and select an operation according to the contents. Thereafter, the control unit 110 proceeds with the process to step S112.
(Step S112) The motion control unit 116 causes the character CR to perform the motion selected in the process of step S110. The voice synthesizer 113 also generates voice information representing abnormality information, translation results, and the like, and outputs voice in cooperation with the action of the character CR. And the control part 110 complete | finishes the process shown in FIG.
The above is the description of the operation of the information processing apparatus 10.

[Summary of First Embodiment]
As described above, the information processing apparatus 10 according to the present embodiment includes the voice recognition unit 111 (an example of an acquisition unit) that acquires first sentence information indicating a sentence in a first language (for example, Japanese), the first A translation unit 112 for executing a translation process for generating second sentence information indicating a sentence translated from a sentence indicated by one sentence information into a second language (for example, English) different from the first language; A second evaluation unit 115 that evaluates the presence or absence, and a motion control unit 116 that moves a character (for example, a character CR) based on the second evaluation by the second evaluation unit 115 are provided.

  Thereby, in the translation process, the information processing apparatus 10 detects an abnormality that lowers the accuracy of translation, and indicates the cause and countermeasure to the user in an easy-to-understand manner, so that the translation can be performed appropriately. Specifically, the information processing apparatus 10 can easily show the user a problem such as the phrase being unsuitable for translation and a countermeasure for each problem.

In addition, the information processing apparatus 10 includes a first evaluation unit 114 that evaluates the presence or absence of abnormality in the speech recognition process that generates the first sentence information from the speech information indicating the speech in the first language. The first sentence information is acquired by executing the voice recognition process, and the motion control unit 116 is based on one or both of the first evaluation by the first evaluation unit and the second evaluation by the second evaluation unit. To move the character.
Thereby, the information processing apparatus 10 detects an abnormality that lowers the accuracy of translation in the speech recognition process, and easily shows the cause and the countermeasure to the user. Therefore, translation can be performed appropriately. Specifically, the information processing apparatus 10 can easily show the user a problem such as an unclear statement or a low voice, and a countermeasure for each problem.

The motion control unit 116 moves the character based on the second sentence information when there is no abnormality in the speech recognition process and there is no abnormality in the translation process.
As a result, when there is no abnormality in the speech recognition process or the translation process, the motion control unit 116 can move the character according to the content of the sentence of the translated word, so that the conversation between users can be performed more smoothly. Can proceed.

The motion control unit 116 moves the character based on the first sentence information when there is no abnormality in the first evaluation and there is an abnormality in the second evaluation.
Thereby, the motion control unit 116 can move the character according to the content of the sentence before translation when there is no abnormality in the speech recognition process and there is an abnormality in the translation process. That is, even if the translated word cannot be output, at least partially, the content of the speech of the speaker SP can be notified to the receiver RE.

In addition, the information processing apparatus 10 is action specification information that defines a correspondence relationship between the character evaluation and / or the first evaluation and / or the second evaluation, and stores different action specification information for each language. An operation information storage unit is provided, and the operation control unit selects an operation corresponding to one or both of the first evaluation and the second evaluation with reference to the operation definition information corresponding to the first language or the second language, Causes the character to perform the selected action.
Thereby, the information processing apparatus 10 can appropriately select a different gesture for each cultural area and operate the character.

In addition, the information processing apparatus 10 includes a scene action storage unit that stores action definition information that is different in each usage scene and that defines a correspondence relationship that is not defined in the action definition information for each language. The section refers to the action definition information according to the usage scene of the own device, selects an action corresponding to one or both of the first evaluation and the second evaluation, and causes the character to perform the selected action.
Thereby, the information processing apparatus 10 can move a character with a different pattern for each use scene. Further, the information processing apparatus 10 can efficiently define the character motion pattern because the motion defined in the motion definition information for each language does not need to be defined in the motion definition information for each use scene. .

In addition, the information processing apparatus 10 receives an operation for selecting an action of a character for which action defining information is set, an operation reception unit 132 that receives an operation, and the language motion information storage unit 122 and the scene motion based on the operation received by the operation reception unit 132. An operation registration unit 117 that stores the operation defining information is provided in either or both of the information storage unit 123.
Thereby, the information processing apparatus 10 can provide communication with a higher degree of freedom in order to allow the user to set the character motion.

[Second Embodiment]
[Outline of information processing equipment]
A second embodiment of the present invention will be described. Here, the same code | symbol is attached | subjected to the structure similar to embodiment mentioned above, and description is used.
Similar to the information processing apparatus 10 according to the first embodiment, the information processing apparatus 10A according to the present embodiment is an apparatus that supports interpretation of conversations in two or more different languages. However, the information processing apparatus 10A causes the character CR to perform an action directed not only to the speaker SP but also to the receiver RE when an abnormality is detected in the speech recognition process or the translation process.

FIG. 8 is a diagram for explaining an overview of the information processing apparatus 10A according to the present embodiment.
FIG. 8 shows three scenes C21 to C23 related to use of the information processing apparatus 10A. In these first to third scenes C21 to C23, voice reception units 131-1 and 131-2, a character CR, a speaker SP, and a receiver RE are shown, respectively.
In the first scene C21, the speaker SP is speaking.
Next, in the second scene C22, the speaker RE (a user different from the speaker SP) starts to speak while the speaker SP is speaking. In this case, since the utterance of the speaker SP and the utterance of the receiver RE are mixed, the accuracy of voice recognition may be lowered, which is not preferable. Therefore, the information processing apparatus 10 </ b> A causes the character CR to perform an action of stopping the speech of the listener RE.
In the third scene thereafter, the listener RE stops speaking by the action of the character CR. As described above, the information processing apparatus 10A can indicate not only the speaker SP but also the receiver RE not only the cause of the translation that is difficult, and the countermeasures, but can appropriately perform the translation. .

[Configuration of information processing device]
Next, the configuration of the information processing apparatus 10A will be described.
FIG. 9 is a block diagram illustrating an example of a functional configuration of the information processing apparatus 10A according to the present embodiment.
The information processing apparatus 10A includes a control unit 110A instead of the control unit 110 included in the information processing apparatus 10 according to the first embodiment. In addition, the control unit 110A includes an operation control unit 116A instead of the operation control unit 116 included in the control unit 110.
Similar to the motion control unit 116, the motion control unit 116A refers to the motion regulation information and selects the motion of the character CR. However, in the action defining information referred to by the action control unit 116A, for example, information on the target person of the action performed by the character CR, such as for the speaker SP and the receiver RE, is described.
The above is the description of the configuration of the information processing apparatus 10A.

[Summary of Second Embodiment]
As described above, in the information processing apparatus 10A according to the present embodiment, the motion control unit 116A causes the character to perform a motion on the listener RE.
Thus, the information processing apparatus 10A causes the character CR to perform an action not only for the speaker SP but also for the receiver RE when an abnormality is detected in the speech recognition process or the translation process. Therefore, the information processing apparatus 10A can prepare an environment in which the speech of the speaker SP is easily translated and can appropriately perform the translation.

[Third embodiment]
[Outline of information processing equipment]
A third embodiment of the present invention will be described. Here, the same code | symbol is attached | subjected to the structure similar to embodiment mentioned above, and description is used.
Similar to the information processing apparatus 10 according to the first embodiment, the information processing apparatus 10B according to the present embodiment is an apparatus that supports interpretation of conversations in two or more different languages. However, the information processing apparatus 10B detects a translation error in accordance with the response of the listener RE to the voice output from the own apparatus.

FIG. 10 is a diagram for explaining an overview of the information processing apparatus 10B according to the present embodiment.
FIG. 10 shows three scenes C31 to C33 related to use of the information processing apparatus 10B. In these first to third scenes C31 to C33, voice receiving units 131-1 and 131-2, a character CR, a speaker SP, and a receiver RE are shown, respectively.
In the first scene C31, the information processing apparatus 10B outputs a voice obtained by translating the speech of the speaker SP.
Next, in the second scene C32, the listener RE cannot hear the speech from the information processing apparatus 10B, and therefore listens again.
In the third scene C33 thereafter, the information processing apparatus 10B performs an operation of showing an apology to the listener RE because there is a possibility that there is an error in the translation of the voice output in the first scene C31. The sound for confirmation is output. In this way, the information processing apparatus 10B can confirm the translation result based on the reaction of the listener RE with respect to the voice output from the own apparatus, and can take appropriate measures for intention transmission.

[Configuration of information processing device]
Next, the configuration of the information processing apparatus 10B will be described.
FIG. 11 is a block diagram illustrating an example of a functional configuration of the information processing apparatus 10B according to the present embodiment.
The information processing apparatus 10B includes a control unit 110B instead of the control unit 110 included in the information processing apparatus 10 according to the first embodiment. Further, the control unit 110B includes an operation control unit 116B instead of the operation control unit 116 included in the control unit 110.

  Similar to the motion control unit 116, the motion control unit 116B refers to the motion regulation information and selects the motion of the character CR. However, the motion control unit 116B is configured such that the listener RE utters a predetermined word or sentence in response to the translated speech output when the accuracy of the first sentence or the second sentence is lower than a predetermined threshold. If a particular action is selected. Here, the predetermined words and sentences are, for example, words and sentences used for listening back. The specific operation is, for example, an operation that shows an apology to the listener RE or an operation that prompts the speaker SP to speak again. Predetermined words and sentences and actions corresponding to these are described in, for example, action definition information.

For example, the motion control unit 116B causes the character CR to perform the specific motion only when the predetermined word or sentence is input during a predetermined period after the voice is output from the voice output unit 14. You may do it. Further, the voice synthesis unit 113 may cause the voice output unit 14 to output a voice corresponding to the operation in accordance with the operation.
The above is the description of the configuration of the information processing apparatus 10B.

[Summary of Third Embodiment]
As described above, in the information processing apparatus 10 </ b> B according to the present embodiment, the motion control unit 116 </ b> B moves the character when a specific voice input is performed on the voice output from the voice output unit 14.
Thereby, the information processing apparatus 10B detects an error in translation according to the reaction of the listener RE with respect to the voice output from the own apparatus.

[Modification]
The embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to the above-described embodiment, and includes a design and the like within a scope not departing from the gist of the present invention. For example, the configurations described in the first to third embodiments described above can be arbitrarily combined or separated.

  For example, the information processing devices 10, 10A, and 10B may be terminal devices connected via a network. In this case, the speaker SP and the listener RE own the information processing apparatuses 10, 10A, and 10B, respectively, and the first sentence information, the second sentence information, the first evaluation information, between the information processing apparatuses 10, 10A, and 10B, The operations of the information processing apparatuses 10, 10A, and 10B according to the above-described embodiments may be realized by transmitting and receiving the second evaluation information and the like. Further, the information processing apparatuses 10, 10A, 10B may be server apparatuses. In this case, each of the speaker SP and the receiver RE may use a terminal device that communicates with the server device and receive a translation service via the character CR displayed on the terminal device.

  The first evaluation unit 114 may perform only a part of the speech recognition evaluation process. In this case, for example, the voice recognition unit 111 measures an evaluation index such as a certainty factor and outputs the measurement result to the first evaluation unit 114. Then, the first evaluation unit 114 compares the acquired measurement result with a predetermined threshold value to determine whether there is an abnormality in the voice recognition process. In another example, the first evaluation unit 114 measures an evaluation index such as a certainty factor and outputs the measurement result to the operation control units 116, 116A, and 116B. Then, the operation control units 116, 116A, and 116B compare the acquired measurement results with a predetermined threshold value to determine whether or not there is an abnormality in the voice recognition process.

Similarly, the second evaluation unit 115 may perform only a part of the translation evaluation process. In this case, for example, the translation unit 112 measures an evaluation index such as a certainty factor and outputs the measurement result to the second evaluation unit 115. Then, the second evaluation unit 115 compares the acquired measurement result with a predetermined threshold value to determine whether there is an abnormality in the translation process. In another example, the second evaluation unit 115 measures an evaluation index such as a certainty factor and outputs the measurement result to the operation control units 116, 116A, and 116B. Then, the operation control units 116, 116A, and 116B compare the acquired measurement results with a predetermined threshold value to determine whether there is an abnormality in the translation process.
As described above, the processing performed by each configuration of the information processing apparatuses 10, 10A, and 10B in the above-described embodiments may be arbitrarily separated or performed by another configuration.

  In the above-described embodiment, various threshold values used for the speech recognition evaluation process and the translation evaluation process may not be fixed. For example, these threshold values may be determined in advance according to the language or usage scene. In this case, these threshold values may be described in advance in the action definition information for each language or the action definition information for each use scene.

  Moreover, although embodiment mentioned above demonstrated the case where the speech by the voice of speaker SP was translated, it is not restricted to this. For example, the information processing apparatus 10 may directly acquire the first sentence information without performing voice recognition. Specifically, the information processing device 10 may include an input device for character input such as a keyboard, and may acquire the first sentence information based on the input to the input device. For example, the information processing apparatus 10 may acquire the first sentence information transmitted from the external apparatus. As described above, the information processing apparatus 10 may include an acquisition unit that acquires the first sentence information through character input, communication with another apparatus, or the like.

  Also, a program for realizing the functions of the information processing apparatuses 10, 10A, and 10B described above is recorded on a computer-readable recording medium, and the program recorded on the recording medium is read into a computer system and executed. Thus, the processing as the information processing devices 10, 10A, 10B may be performed. Here, “loading and executing a program recorded on a recording medium into a computer system” includes installing the program in the computer system.

  The “computer system” here includes an OS and hardware such as peripheral devices. The “computer system” may include a plurality of computer devices connected via a network including a communication line such as the Internet, WAN, LAN, and dedicated line. 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.

  As described above, the recording medium storing the program may be a non-transitory recording medium such as a CD-ROM. The recording medium also includes a recording medium provided inside or outside that is accessible from the distribution server in order to distribute the program. The code of the program stored in the recording medium of the distribution server may be different from the code of the program that can be executed by the terminal device. That is, the format stored in the distribution server is not limited as long as it can be downloaded from the distribution server and installed in a form that can be executed by the terminal device. Note that the program may be divided into a plurality of parts, downloaded at different timings, and combined in the terminal device, or the distribution server that distributes each of the divided programs may be different.

  Furthermore, the “computer-readable recording medium” holds a program for a certain period of time, such as a volatile memory (RAM) inside a computer system that becomes a server or a client when the program is transmitted via a network. Including things. The program may be for realizing a part of the functions described above. Furthermore, what can implement | achieve the function mentioned above in combination with the program already recorded on the computer system, and what is called a difference file (difference program) may be sufficient.

DESCRIPTION OF SYMBOLS 10, 10A, 10B ... Information processing apparatus, 11 ... CPU, 12 ... Storage part, 121 ... Operation information storage part, 13 ... Input part, 131 ... Audio | voice reception part, 132 ... Operation reception part, 14 ... Audio | voice output part, 15 ... Display unit, 110, 110A, 110B ... Control unit, 111 ... Speech recognition unit, 112 ... Translation unit, 113 ... Speech synthesis unit, 116, 116A, 116B ... Operation control unit

Claims (9)

An acquisition unit for acquiring first sentence information indicating a sentence in a first language;
A translation unit for executing a translation process for generating second sentence information indicating a sentence obtained by translating the sentence indicated by the first sentence information into a second language different from the first language;
A second evaluation unit for evaluating the presence or absence of abnormality in the translation process;
A motion control unit that moves the character based on the second evaluation by the second evaluation unit;
An information processing apparatus comprising: A first evaluator that evaluates the presence or absence of an abnormality in the speech recognition process for generating the first sentence information from speech information indicating the speech in the first language;
With
The acquisition unit acquires the first sentence information by executing the voice recognition process,
The information processing apparatus according to claim 1, wherein the motion control unit moves the character based on one or both of a first evaluation by the first evaluation unit and a second evaluation by the second evaluation unit. The information processing apparatus according to claim 2, wherein the motion control unit moves the character based on the second sentence information when there is no abnormality in the speech recognition process and there is no abnormality in the translation process. . The said action control part operates the said character based on said 1st sentence information, when there is no abnormality in the said speech recognition process and there is abnormality in the said translation process. Information processing device. A language action information storage unit that stores action definition information that defines a correspondence relationship between the first evaluation and / or the second evaluation and the action of the character, and that is different from each other for each language,
With
The operation control unit selects and selects an operation corresponding to one or both of the first evaluation and the second evaluation with reference to the operation definition information according to the first language or the second language. The information processing apparatus according to claim 2, wherein the character is caused to perform an action. A scene action storage unit that stores action regulation information that defines the correspondence relationship, which is different action definition information for each use scene and is not defined in the action definition information for each language,
With
The motion control unit selects a motion corresponding to one or both of the first evaluation and the second evaluation with reference to motion regulation information corresponding to a use scene of the device, and selects the selected motion as the character. The information processing apparatus according to claim 5. An operation accepting unit for accepting an operation for setting the operation regulation information;
Based on the operation accepted by the operation accepting unit, an operation registration unit that stores the action defining information in either or both of the language action information storage unit and the scene action storage unit;
The information processing apparatus according to claim 6. A first step in which the information processing apparatus acquires first sentence information indicating a sentence in a first language;
A second step in which the information processing apparatus executes a translation process for generating second sentence information indicating a sentence obtained by translating the sentence indicated by the first sentence information into a second language different from the first language;
A third step in which the information processing apparatus evaluates whether there is an abnormality in the translation process;
A fourth step in which the information processing apparatus moves the character based on the evaluation in the third step;
An information processing method including: On the computer,
A first step of acquiring first sentence information indicating a sentence in a first language;
A second step of executing a translation process for generating second sentence information indicating a sentence obtained by translating the sentence indicated by the first sentence information into a second language different from the first language;
A third step of evaluating the presence or absence of abnormality in the translation process;
A fourth step of moving the character based on the evaluation in the third step;
A program that executes

Images