JP2017120616A - Machine translation method and machine translation system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To further improve function of a machine translation system.SOLUTION: A machine translation method includes: an acquisition step (S1) for acquiring text information before translation that is generated when first voice data representing input voice spoken in a first language is converted into text information; a determination step (S2) for determining whether the text information before translation acquired in the acquisition step (S1) includes first specific text information stored in a storage unit; and an output step (S3) for outputting, when it is determined in the determination step (S2) that the text information before translation includes the first specific text information, at least one of second specific text information that is stored in the storage unit and associated with the first specific text information and second voice data of the second specific text information.SELECTED DRAWING: Figure 6

Description

  The present invention relates to a machine translation method and a machine translation system.

  In recent years, machine translation systems that can be translated into multi-languages and output as speech by speaking in their own language have attracted attention. Such a machine translation system is expected to remove the world language barrier and promote global communication.

  In recent years, the scope of such machine translation systems has been expanded from those for personal use to those for business use. As a communication tool with foreign visitors, it can be used in public facilities and stores. It is being considered.

  When such a machine translation system is introduced for business use, there is an utterance sentence or a conversation sentence that frequently requires translation for each use purpose. Therefore, for example, Patent Document 1 discloses a technique for registering in advance a sentence that is frequently requested for translation in association with a message code. Thereby, the user can call the sentence matched with the message code by designating a desired message code at the time of use.

JP-A-9-139969

  However, the technique disclosed in Patent Document 1 cannot reduce the burden on a speaker as a whole for an utterance sentence and the like that are frequently requested for translation and having different contents for each use in business.

  Therefore, it is necessary to further improve the function of the machine translation system that reduces the burden on the speaker and shortens the translation time of the spoken sentence in an uttered sentence that requires frequent translation. The

  The present invention has been made in view of the above circumstances, and an object thereof is to provide a machine translation method and a machine translation system that can further improve the functions of a machine translation system.

  In order to achieve the above object, a machine translation method according to an aspect of the present disclosure is a machine translation method in a machine translation system, in which a specific word or a specific sentence expressed in a first language is stored in a storage unit. 1st specific text information and 2nd specific text information which shows the fixed sentence which is the word or sentence expressed in the 2nd language which is a language different from the said 1st language prepared beforehand, Comprising: The said specific word or the said specific sentence The second specific text information having no translation equivalence and at least one of the second audio data of the second specific text information is stored in association with each other, and the input speech uttered in the first language is stored. An acquisition step of acquiring pre-translation text information generated by converting first audio data representing text information, and the pre-translation text information acquired in the acquisition step, In the determination step of determining whether or not the first specific text information stored in the storage unit is included, and in the determination step, the pre-translation text information includes the first specific text information. If it is determined that the second specific text information is stored in the storage unit and is associated with the first specific text information, the second audio data of the second specific text information is small. The other includes an output step of outputting.

  In order to achieve the above object, a machine translation system according to an aspect of the present disclosure includes first specific text information indicating a specific word or a specific sentence expressed in a first language, and the first specific text information prepared in advance. Second specific text information indicating a fixed sentence that is a word or sentence expressed in a second language which is a language different from the language, the second specific text information having no translation equivalence with the specific word or the specific sentence, and A storage unit that associates and stores at least one of the second voice data of the second specific text information, and the first voice data representing the input voice uttered in the first language is converted into text information and generated. The acquisition unit for acquiring the pre-translation text information, and the first specific text information stored in the storage unit is included in the pre-translation text information acquired in the acquisition unit. When it is determined by the determination unit that determines whether or not the first specific text information is included in the pre-translation text information by the determination unit, the storage unit stores the first specific text information, and the first An output unit that outputs at least one of the second specific text information associated with the one specific text information and the second audio data of the second specific text information.

  These general or specific aspects may be realized by a system, a method, an integrated circuit, a computer program, or a computer-readable recording medium such as a CD-ROM. The system, method, integrated circuit, computer You may implement | achieve with arbitrary combinations of a program and a recording medium.

  According to the machine translation method and the machine translation system of the present disclosure, it is possible to realize further improvements for improving the function of the machine translation system.

FIG. 1 is a diagram illustrating an example of a functional configuration of a conventional machine translation system. FIG. 2A is a diagram illustrating an overall image of services provided by the information management system according to the present disclosure. FIG. 2B is a diagram illustrating a modification of a partial configuration of the information providing system according to the present disclosure. FIG. 2C is a diagram illustrating a modification of a partial configuration of the information providing system according to the present disclosure. FIG. 3 is a block diagram showing an example of the configuration of the machine translation system in the first embodiment. FIG. 4 is a block diagram illustrating an example of a detailed configuration of the translation determination processing unit illustrated in FIG. FIG. 5 is a block diagram showing another aspect of the configuration of the machine translation system in the first embodiment. FIG. 6 is a flowchart showing an outline of the operation of the machine translation system in the first embodiment. FIG. 7 is a flowchart showing a specific mode of operation of the machine translation system in the first embodiment. FIG. 8 is a diagram illustrating an example of a specific sentence and a fixed sentence stored in association with each other in the storage unit according to the first embodiment. FIG. 9 is a diagram illustrating an example of a usage scene of a machine translation system including a display. FIG. 10 is a diagram illustrating an example of a configuration of a machine translation system according to a modification of the first embodiment. FIG. 11 is a diagram illustrating an example of a configuration of an information terminal according to a modification of the first embodiment. FIG. 12 is a sequence diagram showing an operation of the machine translation system in the modification of the first embodiment. FIG. 13 is a sequence diagram illustrating another example of the operation of the machine translation system in the modification of the first embodiment. FIG. 14 is a flowchart showing a specific mode of operation of the machine translation system in the second embodiment. FIG. 15 is a flowchart showing a specific mode of operation of the machine translation system in the third embodiment. FIG. 16 is a diagram illustrating an example of fixed phrases corresponding to specific words and utterance orders stored in association with each other in the storage unit according to the third embodiment. FIG. 17 is a diagram showing an overview of services provided by the information management system in service type 1 (in-house data center type cloud service). FIG. 18 is a diagram illustrating an overall image of services provided by the information management system in service type 2 (IaaS-based cloud service). FIG. 19 is a diagram illustrating an overall image of services provided by the information management system in service type 3 (PaaS use type cloud service). FIG. 20 is a diagram illustrating an overview of services provided by the information management system in service type 4 (SaaS-based cloud service).

(Knowledge that became the basis of this disclosure)
Machine translation appeared around 1990. However, the initial translation accuracy of machine translation was about 60% for English-Japanese translation and about 50% for Japanese-English translation. Therefore, it is necessary to manually correct a considerable number of errors caused by machine translation, and complete machine translation has been regarded as a dream technology. However, in recent years, translation accuracy has been rapidly improved by using advanced machine learning techniques such as Deep Learning for machine translation. Therefore, machine translation is being used as a translation system that can be easily used in PC applications, Web applications, smartphone applications, and the like.

  On the other hand, the recognition rate of speech recognition has been improved by various technological developments using statistical methods. For this reason, voice recognition is being used as a voice control interface for electrical appliances, such as device control based on utterances, as well as textualization of user utterance content.

  Under these circumstances, a machine translation system that speaks in its own language, translates into multiple languages, and outputs a voice has been attracting attention as a means of removing the world language barrier and promoting global communication.

  FIG. 1 is a diagram illustrating an example of a functional configuration of a conventional machine translation system. A machine translation system 90 shown in FIG. 1 includes a speech input unit 91, a speech recognition unit 92, a translation unit 93, a speech synthesis unit 94, and a speech output unit 95.

  The voice input unit 91 receives a voice spoken by the speaker in the first language. The voice input unit 91 converts the input voice into voice data and outputs the voice data to the voice recognition unit 92. The voice recognition unit 92 executes voice recognition processing on the received voice data, and executes processing for converting the voice data into text data of the first language. The voice recognition unit 92 outputs the converted text data of the first language to the translation unit. The translation unit 93 executes a translation process for translating the received text data in the first language into the second language, and generates text data in the second language. The translation unit 93 outputs the generated text data of the second language to the speech synthesis unit 94. The speech synthesizer 94 converts the received second language text data into second language speech data and outputs the speech data to the speech output unit 95. The voice output unit 95 outputs (speaks) the received second language voice data as second language voice.

  In this way, the machine translation system 90 receives the utterance content of the speaker who spoke in the first language, translates it into the second language, and then outputs the speech to the listener as the second language speech. Thereby, communication between speakers using different languages can be realized.

  Such machine translation systems have been used mainly by individuals such as personal travel destinations and SNS. However, with recent improvements in speech recognition accuracy and translation accuracy, communication tools with foreign visitors As a result, the use for business in public facilities, stores, etc. is also being considered.

  However, when a machine translation system is introduced for business use, more accurate translation, that is, higher translation accuracy is required than that for personal use. On the other hand, for example, for business use in hotels, travel agencies, transportation facilities, information centers, medical fields, shops, etc., there are special conversation sentences such as original conversations, words, and phrases for each use. For special conversational sentences, there is a problem in that speech recognition and translation results are not performed correctly unless the machine translation system performs learning in advance using, for example, machine learning technology.

  Furthermore, there are utterances or conversations that frequently require translation for each business use. In other words, a speaker (service provider) frequently has a sentence (an utterance sentence or a conversation sentence) to be uttered or conversed with a partner (service recipient) whose language is a different language. Therefore, in order for the machine translation system to translate these sentences for the other party (service recipient), the speaker (service provider) needs to utter these sentences many times with the same content. This is a burden on the speaker (service provider).

  In particular, if the length of these sentences is long, not only will the burden on the speaker be further increased, but the speech recognition and translation results of these sentences will not be performed correctly at one time in the machine translation system. It leads to. That is, when these sentences are long, a lot of noise around the utterance environment is input to the machine translation system. Therefore, in the speech recognition processing of these sentences (uttered sentences or conversations). This increases the frequency of occurrence of erroneous recognition and leads to output of an incorrect translated sentence. As a result, the speaker has to re-speak again, which further burdens the speaker.

  As a countermeasure to such a problem, for example, in Patent Document 1, in a message data receiving device that creates a standard message, a sentence that is frequently required to be translated may be registered in advance in association with a message code. It is disclosed.

  More specifically, in Patent Document 1, a correspondence table in which frequently used words or sentences are assigned to respective message codes (No) is stored in the standard message memory (33) in advance. When the message data is included in the received data received by the receiving circuit (22), the message code included in the received data is based on the correspondence table stored in the standard message memory (33). The call contents (words, sentences) corresponding to are extracted. And the message of a received signal is produced | generated by substituting with the message code contained in received data. In this way, a speaker can call a sentence associated with a message code by designating a desired message code at the time of use. Thereby, a long message can be easily generated just by using the message code, and the burden on the user can be reduced.

  However, in the technique disclosed in Patent Document 1, a word or sentence registered in advance in the correspondence table is associated with a numerical value having no meaning such as a number as a message code. For this reason, the user needs to store which number the word or sentence is associated with in the correspondence table, and the burden on the user is large. In particular, the more words or sentences registered in advance, the greater the burden on the user.

  That is, in the technique disclosed in Patent Document 1, in order to reduce the burden on a speaker for an utterance sentence and the like that are frequently required to be translated and have different contents for each use in business. The technical problem is not solved.

  Therefore, in order to improve the function of the machine translation system that reduces the burden on the utterer and shortens the translation time of the uttered sentence in the uttered sentence that frequently requires translation, further function improvement is necessary. .

  In order to achieve the above object, a machine translation method according to an aspect of the present disclosure is a machine translation method in a machine translation system, in which a specific word or a specific sentence expressed in a first language is displayed in a storage unit. 1st specific text information and 2nd specific text information which shows the fixed sentence which is the word or sentence expressed in the 2nd language which is a language different from the said 1st language prepared beforehand, Comprising: The said specific word or the said specific sentence The second specific text information having no translation equivalence and at least one of the second audio data of the second specific text information is stored in association with each other, and the input speech uttered in the first language is stored. An acquisition step of acquiring pre-translation text information generated by converting first audio data representing text information, and the pre-translation text information acquired in the acquisition step, In the determination step of determining whether or not the first specific text information stored in the storage unit is included, and in the determination step, the pre-translation text information includes the first specific text information. If it is determined that the second specific text information is stored in the storage unit and is associated with the first specific text information, the second audio data of the second specific text information is small. The other includes an output step of outputting.

  As a result, the utterer does not utter all sentences of the first language utterances that are frequently used, but only utters the words including the specific words or specific sentences (first specific text information) of the first language. Since one of the fixed phrases (second specific text information) and the voice data of the second language frequently used can be output, the utterance burden on the speaker is reduced. The second language fixed sentence (second specific text information) and the voice data thereof are associated with a word or the like that briefly expresses the second language fixed sentence in the first language. That is, the second specific text information and the voice data thereof according to the first embodiment are not associated with numbers having no meaning. Thereby, it is possible to reduce a burden that the user has to store the association in advance or separately.

  Further, for example, in the determination step, it is determined whether or not the pre-translation text information matches the first specific text information stored in the storage unit. In the output step, in the determination step When it is determined that the pre-translation text information and the first specific text information match, the second specific text information stored in the storage unit and associated with the first specific text information and At least one of the second audio data of the second specific text information may be output.

  Thereby, only when the content of the utterance sentence of the speaker (text information before translation) matches the first specific text information, one of the second specific text information and its voice data can be output.

  That is, when the speaker utters only the first specific text information, one of the second specific text information and the voice data can be output as a translation result. On the other hand, when the utterer utters an utterance including a phrase other than the first specific text information, a translated sentence (post-translation text information) in which the uttered content is translated as it is can be output. Thus, the speaker can use the first specific text information as a word or sentence in conversation.

  Also, for example, in the storage unit, the second specific text information includes an order in which each of the two or more first specific text information and the two or more first specific text information should appear in a sentence. The order information shown is stored in association with each other, and in the determination step, the two or more first specific text information stored in the storage unit is added to the pre-translation text information acquired in the acquisition step. It is determined whether or not the two or more first specific text information is included in the order indicated by the order information. In the output step, the pre-translation text information in the determination step includes: The two or more first specific text information stored in the storage unit is included, and the two or more first specific text information is the previous When it is determined that they are included in the order indicated by the order information, the second specific text information corresponding to the two or more first specific text information and the order information, and the second specific text information corresponding to the second specific text information. At least one of the audio data may be output.

  As a result, when the plurality of first specific text information appear in a predetermined order in the content of the utterance sentence of the speaker (pre-translation text information), the second specification associated with the plurality of first specific text information One of text information and its voice data can be output. That is, the speaker can output either the second specific text information corresponding to the utterance sentence including the first specific text information and the voice data, or the translation sentence in which the uttered content is translated as it is. It can be selected as appropriate according to the manner of utterance.

  Further, for example, in the storage unit, the one or more first specific text information that is one or more different first specific text information, each including the same specific word and indicating a different specific sentence. The specific text information may be stored in association with each other.

  Thereby, a plurality of different phrases and a plurality of different keywords can be set for one second specific text information. Thereby, when the user wants to output the second specific text information, the second specific text information can be output by any of a plurality of phrases or a plurality of keywords.

  Further, for example, in the output step, when it is determined in the determination step that the first specific text information is not included in the pre-translation text information, the pre-translation text information is translated into the second language. The post-translation text information may be output.

  Thereby, when the first specific text information is included in the pre-translation text information, the translation process for translating the pre-translation text information into the second language is omitted, but the first specific text information is included in the pre-translation text information. Is not included, translation processing for translating the pre-translation text information into the second language is executed. Thereby, it is possible to shorten the translation processing time for frequently used utterances or original conversations, words, and phrases.

  For example, in the output step, when it is determined in the determination step that the first specific text information stored in the storage unit is included in the pre-translation text information, the post-translation text information is It may not be output.

  As described above, when the first specific text information is included in the pre-translation text information, the translation processing for translating the pre-translation text information into the second language is not executed. Accordingly, the translation processing in the machine translation processing system can be simplified and the translation processing capability can be used for other processing, so that the function of the machine translation system can be improved.

  In addition, for example, in the storage unit, the third specific text information obtained by translating the second specific text information into the first language is at least the first specific text information and the second specific text information. Stored in association with second specific text information, and in the output step, when outputting at least one of the second specific text information and the second audio data of the second specific text information, The third specific text information may be output.

  As described above, when outputting the fixed sentence of the second language, which is the second specific text information, the sentence of the first language that is in translation relation with the fixed sentence of the second language is also output. Thereby, the speaker can confirm what kind of information is transmitted to the listener based on the utterance sentence uttered by the speaker.

  Further, for example, the third specific text information output in the output step may be displayed on a display.

  Thus, the speaker can visually confirm what kind of information is transmitted to the listener based on the utterance sentence uttered by the speaker.

  Further, for example, the machine translation system is connected to an information terminal having a display by a predetermined communication means, and in the output step, the second specific text information and the second sound of the second specific text information are included. At least one of the data may be output to the information terminal by the predetermined communication means.

  In addition, for example, when the second specific text information is output in the output step, the information terminal performs voice synthesis processing on the second specific text information to generate second voice data, and generates the second The voice indicated by the voice data may be output.

  For example, the machine translation method may be used in a predetermined situation between a person who speaks the first language and a person who speaks the second language.

  In order to achieve the above object, a machine translation system according to an aspect of the present disclosure includes first specific text information indicating a specific word or a specific sentence expressed in a first language, and the first specific text information prepared in advance. Second specific text information indicating a fixed sentence that is a word or sentence expressed in a second language which is a language different from the language, the second specific text information having no translation equivalence with the specific word or the specific sentence, and A storage unit that associates and stores at least one of the second voice data of the second specific text information, and the first voice data representing the input voice uttered in the first language is converted into text information and generated. The acquisition unit for acquiring the pre-translation text information, and the first specific text information stored in the storage unit is included in the pre-translation text information acquired in the acquisition unit. When it is determined by the determination unit that determines whether or not the first specific text information is included in the pre-translation text information by the determination unit, the storage unit stores the first specific text information, and the first An output unit that outputs at least one of the second specific text information associated with the one specific text information and the second audio data of the second specific text information.

  These general or specific aspects may be realized by a system, a method, an integrated circuit, a computer program, or a computer-readable recording medium such as a CD-ROM. The system, method, integrated circuit, computer You may implement | achieve with arbitrary combinations of a program or a recording medium.

  Hereinafter, a machine translation method and the like according to an aspect of the present disclosure will be specifically described with reference to the drawings. Note that each of the embodiments described below shows a specific example of the present invention. Numerical values, shapes, components, steps, order of steps and the like shown in the following embodiments are merely examples, and are not intended to limit the present invention. In addition, among the constituent elements in the following embodiments, constituent elements that are not described in the independent claims indicating the highest concept are described as optional constituent elements. In all the embodiments, the contents can be combined.

(Overview of services provided)
First, an overview of services that provide the machine translation system in the present embodiment as an information management system will be described.

  FIG. 2A is a diagram illustrating an overall image of services provided by the information management system according to the present disclosure. FIG. 2B is a diagram illustrating a modification of a partial configuration of the information providing system according to the present disclosure. FIG. 2C is a diagram illustrating a modification of a partial configuration of the information providing system according to the present disclosure. The information management system shown in FIG. 2A includes a group 11000, a data center operating company 11100, and a service provider 11200.

  The group 11000 is, for example, a company, an organization, a household, etc., and its size does not matter. The group 11000 includes a plurality of devices 11010 and a home gateway 11020 including a first device and a second device. The plurality of devices 11010 include devices that can be connected to the Internet (for example, smartphones, personal computers (PCs) or televisions), and devices that cannot be connected to the Internet by themselves (for example, lighting, washing machines, refrigerators, and the like). Including. The plurality of devices 11010 may include devices that can be connected to the Internet via the home gateway 11020 even if they cannot be connected to the Internet. In addition, the user 10100 uses a plurality of devices 11010 in the group 11000.

  The data center operating company 11100 includes a cloud server 11110. The cloud server 11110 is a virtualization server that cooperates with various devices via the Internet. The cloud server 11110 mainly manages huge data (big data) that is difficult to handle with a normal database management tool or the like. The data center operating company 11100 performs management of data, management of the cloud server 11110, operation of the data center that performs them, and the like. Details of services performed by the data center operating company 11100 will be described later.

  Here, the data center operating company 11100 is not limited to a company that only manages data or manages the cloud server 11110. For example, as shown in FIG. 2B, when a device manufacturer that develops or manufactures one of a plurality of devices 11010 performs data management or cloud server 11110 management, the device manufacturer This corresponds to the data center operating company 11100. Further, the data center operating company 11100 is not limited to one company. For example, as shown in FIG. 2C, when the device manufacturer and the management company jointly or share the management of the data or the cloud server 11110, both or one corresponds to the data center operating company 11100. .

  The service provider 11200 includes a server 11210. The server 11210 referred to here includes, for example, a memory in a personal PC regardless of the scale. In addition, the service provider 11200 may not include the server 11210.

  In the information management system described above, home gateway 11020 is not essential. For example, when the cloud server 11110 performs all data management, the home gateway 11020 is unnecessary. In addition, there may be no device that cannot be connected to the Internet by itself, as in the case where all devices in the home are connected to the Internet.

  Next, the flow of information in the information management system will be described.

  First, the first device or the second device of the group 11000 transmits each log information to the cloud server 11110 of the data center operating company 11100. The cloud server 11110 accumulates log information of the first device or the second device (arrow 11310 in FIG. 2A). Here, the log information is information indicating, for example, driving conditions or operation dates / times of the plurality of devices 11010. For example, the log information includes TV viewing history, recording recording information of the recorder, operation date / time of the washing machine, amount of laundry, opening / closing date / time of the refrigerator, or opening / closing frequency of the refrigerator, but is not limited thereto. Various information that can be acquired from various devices may be included. Note that the log information may be provided directly to the cloud server 11110 from a plurality of devices 11010 themselves via the Internet. Further, the log information may be once accumulated in the home gateway 11020 from a plurality of devices 11010 and provided to the cloud server 11110 from the home gateway 11020.

  Next, the cloud server 11110 of the data center operating company 11100 provides the collected log information to the service provider 11200 in a certain unit. Here, the fixed unit may be a unit that can organize and provide the information collected by the data center operating company 11100 to the service provider 11200, or may be a unit that the service provider 11200 requests. In addition, although the information is provided in a fixed unit, it may not be a fixed unit, and the amount of information to be provided may change depending on the situation. The log information is stored in the server 11210 held by the service provider 11200 as necessary (arrow 11320 in FIG. 2A).

  Then, the service provider 11200 organizes the log information into information suitable for the service provided to the user and provides it to the user. The user to whom the information is provided may be a user 10100 using a plurality of devices 11010 or an external user 10200. As a method of providing information to the users 10100 and 10200, for example, information may be provided directly from the service provider 11200 to the users 10100 and 10200 (arrows 11330 and 11340 in FIG. 2A). As a method for providing information to the user 10100, for example, information may be provided to the user 10100 via the cloud server 11110 of the data center operating company 11100 again (arrows 11350 and 11360 in FIG. 2A). Further, the cloud server 11110 of the data center operating company 11100 may organize the log information into information suitable for the service provided to the user and provide the information to the service provider 11200.

  Note that the user 10100 may be the same as or different from the user 10200.

(Embodiment 1)
Hereinafter, a machine translation system according to the present disclosure will be described.

[Configuration of machine translation system]
FIG. 3 is a block diagram illustrating an example of the configuration of the machine translation system 10 according to the first embodiment.

  The machine translation system 10 is used in a predetermined situation between a person who speaks a first language and a person who speaks a second language. More specifically, as described above, the machine translation system 10 communicates with visiting foreigners in business (predetermined situations) in, for example, hotels, travel agencies, transportation facilities, information centers, medical fields, shops, etc. Used as a tool. As shown in FIG. 3, the machine translation system 10 includes a speech input unit 11, a speech recognition unit 12, a translation unit 13, a speech synthesis unit 14, a speech output unit 15, and a translation determination processing unit 16. Prepare. These may be connected by, for example, an LSI internal bus.

<Voice input unit 11>
The voice input unit 11 receives a voice spoken by the speaker in the first language. The voice input unit 11 converts the input voice into voice data (hereinafter referred to as “pre-translation voice data”) and outputs the voice data to the voice recognition unit 12. In the present embodiment, the voice input unit 11 is composed of a microphone, for example.

<Voice recognition unit 12>
The speech recognition unit 12 performs speech recognition processing on the acquired pre-translation speech data, and converts the pre-translation speech data into text data of the first language (hereinafter referred to as pre-translation text information). Execute. The speech recognition unit 12 outputs the converted pre-translation text information to the translation determination processing unit 16.

  The voice recognition unit 12 is realized by a computer including a CPU, a memory, and the like, and the voice recognition unit 12 may execute the voice recognition process by itself, but is not limited thereto. The voice recognition unit 12 has a communication function and a memory function, and may communicate with the cloud (server group) by predetermined communication means such as the Internet. In this case, the speech recognition unit 12 may transmit the acquired pre-translational speech data to the cloud and acquire the result of executing speech recognition processing on the pre-translational speech data from the cloud.

<Translation unit 13>
When acquiring the pre-translation text information from the translation determination processing unit 16, the translation unit 13 performs a translation process for translating the acquired pre-translation text information into the second language, and the second language text data (hereinafter referred to as the text data) , Post-translation text information and description). The translation unit 13 outputs the generated post-translation text information to the speech synthesis unit 14.

  The translation unit 13 outputs the generated post-translation text information to the speech synthesis unit 14, but is not limited thereto, and outputs the generated post-translation text information to the translation determination processing unit 16. May be output to the speech synthesizer 14. That is, the translation unit 13 may output the generated post-translation text information to the speech synthesis unit 14 via the translation determination processing unit 16.

  Moreover, the translation part 13 is implement | achieved by the computer comprised from CPU, memory, etc., and the translation part 13 may perform a translation process itself, but it is not restricted to it. The translation unit 13 has a communication function and a memory function, and may communicate with the cloud by a predetermined communication means such as the Internet. In this case, the translation part 13 may acquire the result of having performed the translation process with respect to the text information before translation from the cloud which transmits the acquired text information before translation to a cloud.

<Speech synthesizer 14>
When the post-translation text information is acquired, the speech synthesis unit 14 performs speech synthesis processing for generating second language speech data (hereinafter also referred to as post-translation speech data) from the post-translation text information. In addition, when the second specific text information is acquired, the speech synthesizer 14 performs a speech synthesis process for generating second language speech data (hereinafter also referred to as post-translational speech data) from the second specific text information. Then, the speech synthesizer 14 outputs the generated post-translational speech data to the speech output unit 15.

  Note that the speech synthesizer 14 is realized by a computer including a CPU, a memory, and the like, and the speech synthesizer 14 may execute speech synthesis processing by itself, but is not limited thereto. The speech synthesizer 14 may have a communication function and a memory function, communicate with the cloud by a predetermined communication means such as the Internet, and transmit the acquired post-translation text data or second specific text information to the cloud. And the result of performing the speech recognition process with respect to the transmitted post-translation text data or 2nd specific text information may be acquired from a cloud.

<Audio output unit 15>
The voice output unit 15 performs voice output processing for outputting (speaking) the received second language voice data as second language voice. In the present embodiment, the audio output unit 15 is composed of a speaker, for example.

<Translation determination processing unit 16>
FIG. 4 is a block diagram illustrating an example of a detailed configuration of the translation determination processing unit 16 illustrated in FIG.

  For example, as illustrated in FIG. 4, the translation determination processing unit 16 includes an acquisition unit 161, a determination unit 162, a storage unit 163, and an output unit 164.

  The storage unit 163 is first specific text information indicating a specific word or specific sentence expressed in the first language, and a word or sentence expressed in a second language that is a language different from the first language prepared in advance. Second specific text information indicating a fixed phrase, a specific word or specific sentence, and second specific text information having no translation equivalence are stored in association with each other. Here, the first specific text information is, for example, a short sentence (specific sentence) including keywords (specific words) such as “cigarette” and “translator” such as “cigarette explanation” and “translator explanation”. However, only a keyword (specific word) may be used. Also, lack of translation equivalence means that the translation is asymmetric. In other words, instead of a specific sentence as an explanation of a translator, a sentence without a specific sentence or a sentence containing a specific sentence as it is, “Please come. This is a translator. Speak in short English after a beep. It means that it corresponds to the standard sentence "I will translate it into Japanese."

  In the storage unit 163, one second specific text information corresponding to one or more first specific text information that is different one or more first specific text information and includes different specific sentences and includes the same specific word. It may also be stored. That is, the second specific text information may be associated with a plurality of different phrases or a plurality of different keywords. As a result, the second specific text information that is a fixed sentence can be called with any of a plurality of different phrases or a plurality of different keywords.

  The acquisition unit 161 acquires pre-translation text information generated by converting first speech data representing input speech uttered in a first language into text information.

  The determination unit 162 determines whether or not the first specific text information stored in the storage unit 163 is included in the pre-translation text information acquired by the acquisition unit 161.

  If the determination unit 162 determines that the first specific text information is included in the pre-translation text information, the determination unit 162 stores the output unit 164 in the storage unit 163 and associates it with the first specific text information. The second specific text information is output. Here, when the determination unit 162 determines that the pre-translation text information includes the first specific text information stored in the storage unit 163, the post-translation text information acquired by the acquisition unit 161 is output to the output unit 164. Is not output.

  Further, when determining that the first specific text information is not included in the pre-translation text information, the determination unit 162 causes the output unit 164 to output the pre-translation text information.

  The output unit 164 outputs the second specific text information or the pre-translation text information according to the determination result of the determination unit 162. In the present embodiment, the output unit 164 outputs the second specific text information to the speech synthesis unit 14 and the pre-translation text information to the translation unit 13 according to the determination result of the determination unit 162.

  The translation determination processing unit 16 is not limited to the example illustrated in FIG. Hereinafter, this case will be described with reference to FIG. FIG. 5 is a block diagram illustrating another aspect of the configuration of the machine translation system 10 according to the first embodiment. Elements similar to those in FIGS. 3 and 4 are denoted by the same reference numerals, and detailed description thereof is omitted.

  The machine translation system 10A shown in FIG. 5 differs from the machine translation system 10 shown in FIG. 3 in the configuration of the translation determination processing unit 16A. Compared with the translation determination processing unit 16 shown in FIG. 4, the translation determination processing unit 16A does not have the acquisition unit 161 and the output unit 164, is added with a display 17, and the configuration of the determination unit 162A is different.

  The determination unit 162A has all the functions of the determination unit 162 described above, and further includes the functions of the acquisition unit 161 and the output unit 164. That is, the determination unit 162A acquires the pre-translation text information output from the speech recognition unit 12, and determines whether or not the acquired pre-translation text information includes the first specific text information stored in the storage unit 163. To do. If it is not included in the acquired pre-translation text information, the determination unit 162A outputs the pre-translation text information to the translation unit 13.

  On the other hand, when the acquired pre-translation text information includes the first specific text information stored in the storage unit 163, the determination unit 162A does not output the pre-translation text information to the translation unit 13. The determination unit 162A extracts the second specific text information corresponding to the first specific text information determined to be included in the pre-translation text information from the storage unit 163, and outputs the second specific text information to the speech synthesis unit 14.

  For example, the display 17 may display the second specific text information output to the speech synthesizer 14.

  In the storage unit 163, the third specific text information obtained by translating the second specific text information into the first language corresponds to at least the second specific text information among the first specific text information and the second specific text information. It may also be stored.

  In this case, the determination unit 162A outputs the determined first specific text information or the third specific text information corresponding to the extracted second specific text information to the display 17, and the output third specific text information is the display 17 May be displayed. Further, before and after the output of the second language voice data generated from the second text information, the first language voice data generated from the third specific text information corresponding to the second specific text is output separately. It may be.

  Thus, the speaker can visually or auditorily confirm what kind of information is transmitted to the listener based on the utterance sentence uttered by the speaker.

[Operation of Machine Translation System 10]
An outline of the operation of the machine translation system 10 configured as described above will be described.

  FIG. 6 is a flowchart showing an outline of the operation of the machine translation system 10 according to the first embodiment.

  First, the machine translation system 10 performs an acquisition process of acquiring pre-translation text information generated by converting first speech data representing input speech uttered in a first language into text information (S1).

  Next, the machine translation system 10 performs a determination process for determining whether or not the first specific text information stored in the storage unit 163 is included in the pre-translation text information acquired in S1 (S2).

  Next, when it is determined in S2 that the first specific text information is included in the pre-translation text information, the machine translation system 10 stores the first specific text information in the storage unit 163. The output process which outputs the 2nd specific text information matched is performed (S3).

  Next, a specific mode of operation of the machine translation system 10 will be described with reference to FIGS. 7 and 8.

  FIG. 7 is a flowchart showing a specific mode of operation of machine translation system 10 in the first exemplary embodiment.

  As shown in FIG. 7, first, the machine translation system 10 recognizes a speaker's speech, that is, when a speech spoken by the speaker is input to the speech input unit 11 (Y in S11), the speech input unit 11 inputs the speech. The converted voice is converted into pre-translation voice data and output to the voice recognition unit 12.

  Next, the speech recognition unit 12 performs speech recognition processing on the acquired pre-translation speech data (S12), and converts the pre-translation speech data into pre-translation text information in the first language. The speech recognition unit 12 outputs the converted pre-translation text information to the translation determination processing unit 16.

  Next, the translation determination processing unit 16 performs determination processing. That is, the translation determination processing unit 16 determines whether or not the acquired pre-translation text information includes a specific word or specific sentence registered in advance in the storage unit 163 (S13).

  In S13, when it is determined that the specific word or the specific sentence is included in the pre-translation text information (Y in S13), it is stored in the storage unit 163 and associated with the first specific text information. The second specific text information is extracted (S14). Then, the translation determination processing unit 16 outputs the extracted second specific text information to the speech synthesis unit 14. The translation determination processing unit 16 may output the third specific text information corresponding to the extracted second specific text information to a display included in the machine translation system 10.

  On the other hand, when it is determined in S13 that the specific word or specific sentence is not included in the pre-translation text information (N in S13), the pre-translation text information is output to the translation unit 13. The translation unit 13 performs a translation process for translating the acquired pre-translation text information into the second language (S15), and generates post-translation text data. The translation unit 13 outputs the generated post-translation text information to the speech synthesis unit 14.

  Next, the speech synthesizer 14 performs speech synthesis processing for generating post-translational speech data in the second language from the acquired second specific text information or post-translation text information (S16). The speech synthesizer 14 outputs the generated post-translational speech data to the speech output unit 15.

  Next, the speech output unit 15 performs speech output processing for outputting (speaking) the acquired translated speech data as speech in the second language (S17).

  FIG. 8 is a diagram illustrating an example of a specific sentence and a fixed sentence stored in association with each other in the storage unit 163 according to the first embodiment.

  In FIG. 8, a Japanese specific sentence is shown as an example of the first specific text information, and an English fixed sentence is shown as an example of the second specific text information. Also, a fixed sentence translated sentence obtained by translating an English fixed sentence into Japanese is shown as an example of the third specific text information.

  For example, when a speaker utters “I will explain the translator” in Japanese to the machine translation system 10, the machine translation system 10 converts the voice data of the speech uttered by the speaker into Japanese by voice recognition processing. Is converted to pre-translation text information. Next, in the machine translation system 10, a determination process is performed by the translation determination processing unit 16, and the pre-translation text information includes the “description of the translator” that is a specific sentence stored in the storage unit 163. Determine. In this case, the machine translation system 10 does not perform a translation process for translating the Japanese sentence “I will explain the translator” into English, which is spoken by the speaker. Instead, the standard English sentence “Welcome. This is a translation machine. Please speak with a short English after the sound a beep.” Is stored in the storage unit 163 and corresponds to the “description of the translator” which is a specific sentence. "It translated into Japanese." Is extracted as the second specific text information, and speech synthesis processing and speech output processing are executed to output the above-mentioned English fixed sentence speech.

  Here, an example will be described in which a speaker performs an utterance different from the above using the word “translator” and is translated by the machine translation system 10. For example, when a speaker speaks in Japanese to the machine translation system 10 “translators are sold there”, the machine translation system 10 performs speech recognition processing on the voice data of the speech spoken by the speaker. Is converted into Japanese text information before translation. Next, in the machine translation system 10, the translation determination processing unit 16 performs a determination process, and determines that the specific sentence stored in the storage unit 163 is not included in the pre-translation text information. In this case, the machine translation system 10 performs a translation process for translating Japanese sentences “translators are sold there” into English, for example, “We sell a translation machine over there. ”Is translated into English sentences, voice synthesis processing and voice output processing are executed, and the voices of the above English sentences are output.

  The same applies to the case where the speaker utters the machine translation system 10 using the word “cigarette”, and a description thereof is omitted here.

  However, in FIG. 8, the same fixed phrase is associated with “cigarette restriction” and “cigarette description” which are different specific sentences including the same word “cigarette” and stored in the storage unit 163. It has been. As described above, a plurality of different specific sentences (first specific text information) may be associated with the same fixed sentence (second specific text information). Also, a plurality of different specific words (first specific text information) may be associated with the same fixed sentence (second specific text information). Thereby, when the user wants to output the second specific text information, the second specific text information can be output by any of a plurality of phrases or a plurality of keywords.

  The machine translation system 10 may include a display as described above. In this case, the machine translation system 10 performs a determination process, and outputs a standard English sentence stored in the storage unit 163 and corresponding to the specific sentence to a listener whose native language is English. A fixed sentence translation (third specific text information) corresponding to the fixed sentence is displayed to a speaker whose native language is Japanese. Accordingly, when outputting the voice of the fixed sentence (second specific text information) of the second language that is different from the specific sentence of the first language spoken by the speaker but is associated, what kind of fixed sentence is to be output It is possible to inform the speaker whether the sound is output. Also, even when the speaker does not memorize the contents of the fixed sentence, the contents of the fixed sentence in the second language can be confirmed on the spot with the fixed sentence translated sentence (the fixed sentence in the first language). The conversation can be made smoothly.

[Effects]
As described above, according to the first embodiment, the speaker (speaker) only utters the machine translation system 10 including a specific word or specific sentence (first specific text information) in the first language. Thus, it is possible to output a fixed sentence (second specific text information) of the second language that is frequently used, such as an original conversation, a word, or a phrase for each use purpose. Thereby, the utterance burden of the speaker is reduced.

  The second language fixed sentence (second specific text information) is associated with a word or the like that briefly expresses the second language fixed sentence in the first language. That is, the second specific text information of the first embodiment is not associated with a meaningless number or the like. Thereby, it is possible to reduce a burden that the user has to store the association in advance or separately.

  Note that the machine translation system 10 according to the first embodiment may include a display as described above. FIG. 9 is a diagram illustrating an example of a use scene (use case) of the machine translation system 10 including a display. The speaker 50 is a person who speaks the first language and corresponds to a service provider. The listener 51 is a person who speaks the second language and corresponds to a service beneficiary. FIG. 9 shows a scene in which a speaker 50 uses a plurality of languages for a listener 51 at a customer service or the like.

  In this case, when the machine translation system 10 performs the determination process and outputs the second specific text information corresponding to the first specific text information to the listener 51 whose first language is the native language, You may display the 3rd specific text information which is the translation of the 1st language of 1st specific text information. Thereby, it is possible to inform the speaker 50 of what kind of content the second specific text information is being output. Further, even when the speaker 50 does not memorize the content of the second specific text information, the content of the second specific text information can be confirmed on the spot with the third specific text information translated in the first language. The conversation with the listener 51 can be smoothly proceeded.

  In the first embodiment described above, a case has been described in which a voice spoken by a speaker in the first language is input to the machine translation system 10 and the machine translation system 10 outputs a second language voice. However, the present invention is not limited thereto. Absent. Voices spoken by the speaker in the first language may be input to the connected information terminal by the machine translation system and predetermined communication means, and the information terminal may output the second language voice.

(Modification)
For example, each configuration of the machine translation system 10 illustrated in FIG. 3 described in the first embodiment may be distributed to an information terminal and a plurality of servers. Hereinafter, the example in this case is demonstrated as a modification.

[Configuration of Machine Translation System 10A]
FIG. 10 is a diagram illustrating an example of a configuration of a machine translation system 10A according to a modification of the first embodiment. The machine translation system 10A is connected to the information terminal 20 having a display by a predetermined communication unit, and outputs the second specific text information to the information terminal 20 by a predetermined communication unit. As illustrated in FIG. 10, the machine translation system 10 </ b> A includes an information terminal 20, a server 41, a server 42, a server 43, and a server 44. The information terminal 20, the server 41, the server 42, the server 43, and the server 44 are connected by the communication means 30.

<Communication means 30>
The communication means 30 is, for example, a wired or wireless network that is connected to the Internet via an optical line, ADSL, or the like. In this case, the information terminal 20 may be a dedicated terminal, and the server 41, the server 42, the server 43, and the server 44 may be, for example, a cloud server.

  The communication unit 30 is not limited to this, and may be a mobile phone network such as a 3G, 4G, or 5G line. In this case, the information terminal 20 may be a dedicated terminal, and the server 41, the server 42, the server 43, and the server 44 may be, for example, a cloud server.

  Further, the communication means 30 may be a short-range radio such as Bluetooth (registered trademark), Ibeacon, IrDA, Wi-Fi, TransferJet, or extra small radio. In this case, the information terminal 20 may be, for example, a terminal, and the server 41, the server 42, the server 43, and the server 44 may be, for example, a dedicated local server or an on-premises dedicated server.

  Further, the communication means 30 may be a dedicated line with 1-to-N connection. In this case, the information terminal 20 may be, for example, a terminal, and the server 41, the server 42, the server 43, and the server 44 may be, for example, a dedicated local server or an on-premises dedicated server.

  The communication means 30 may be a high-speed wireless network such as DCM (Data Communication Module). In this case, the information terminal 20 may be an in-vehicle terminal, for example, and the server 41, the server 42, the server 43, and the server 44 may be a cloud server, for example.

<Information terminal 20>
FIG. 11 is a diagram illustrating an example of the configuration of the information terminal 20 according to the modification of the first embodiment. Elements similar to those in FIG. 3 are denoted by the same reference numerals, and detailed description thereof is omitted.

  The information terminal 20 receives the voice spoken by the speaker in the first language, and the second specific text information based on the spoken sentence indicated by the inputted voice or the speech sentence is translated into the second language for the listener. Output the translated text. In this way, the information terminal 20 serves as a user interface in the machine translation system 10A.

  As shown in FIG. 11, the information terminal 20 includes a voice input unit 11, a voice output unit 15, a communication unit 21, a storage unit 22, and a display 23. The display 23 is not an essential component. The communication unit 21 is realized by a computer including a CPU, a memory, and the like, and transmits and receives data by the server 41, the server 42, the server 43, and the server 44 and the communication unit 30. The storage unit 22 communicates with the server 41, the server 42, the server 43, and the server 44 by the communication unit 21, and stores data that is acquired or output. For example, when the communication unit 21 acquires the second specific text information or the third specific text information, the display 23 displays the second specific text information or the third specific text information.

  In this modification, the information terminal 20 converts the input voice into pre-translation voice data and transmits it to the server 41 via the communication means 30 when the voice indicating the utterance of the first language is input from the speaker. . The information terminal 20 receives from the server 41 the pre-translation text information in the first language converted by the speech recognition process.

  In addition, the information terminal 20 transmits the pre-translation text information received from the server 41 to the server 42. The information terminal 20 receives the second specific text information or the pre-translation text information from the server 42.

  When receiving the pre-translation text information from the server 42, the information terminal 20 transmits it to the server 43. The information terminal 20 receives post-translation text information from the server 43. The pre-translation text information may be transmitted directly to the server 42 by the server 41 without going through the information terminal 20.

  In addition, when the information terminal 20 receives the second specific text information from the server 42, the information terminal 20 transmits the second specific text information to the server 44. On the other hand, when the translated text information is received from the server 43, the information terminal 20 transmits the translated text information to the server 44. Further, the second specific text information may be transmitted directly to the server 44 by the server 42 without going through the information terminal 20. Then, the information terminal 20 receives the post-translation voice data of the second specific text information or the pre-translation text information from the server 44.

  When the information terminal 20 receives the post-translational voice data of the second specific text information or the pre-translation text information, the information terminal 20 performs voice output by the voice output unit 15.

  The information terminal 20 may include the speech synthesis processing unit 140. In this case, when the information terminal 20 receives the second specific text information from the server 42, the information terminal 20 performs a speech synthesis process on the second specific text information to generate second voice data, and the voice indicated by the generated second voice data Is good as output.

<Server 41 to Server 44>
The server 41 includes a communication unit (not shown) and a voice recognition processing unit 110. The server 41 performs speech recognition processing on the speech data before translation transmitted from the information terminal 20 by the speech recognition processing unit 110, and converts speech data before translation into text information before translation in the first language. Execute the process. In addition, the server 41 transmits the pre-translation text information in the first language to the information terminal 20.

  Note that the server 41 may transmit the pre-translation text information directly to the server 42 without going through the information terminal 20.

  The server 42 includes a communication unit (not shown) and a translation determination processing unit 16A (not shown), that is, a communication unit (not shown), a determination unit 162, and a storage unit 163. Since the determination unit 162 and the storage unit 163 are as described above, detailed description thereof is omitted here.

  When the determination unit 162 determines that the pre-translation text information transmitted from the information terminal 20 includes the first specific text information, the server 42 stores the first specific text information and stores the first specific text information. The second specific text information associated with the text information is transmitted to the information terminal 20. On the other hand, when the determination unit 162 determines that the first specific text information is not included in the pre-translation text information, the server 42 transmits the pre-translation text information to the information terminal 20.

  Note that the server 42 may directly transmit the second specific text information to the server 44 or directly transmit the pre-translation text information to the server 43 without going through the information terminal 20.

  The server 43 includes a communication unit (not shown) and a translation processing unit 130. The server 43 executes a translation process in which the translation processing unit 130 translates the pre-translation text information transmitted from the information terminal 20 into the second language, and generates post-translation text information in the second language. The server 43 transmits the generated post-translation text information to the information terminal 20.

  Note that the server 43 may transmit the post-translation text information directly to the server 44 without going through the information terminal 20.

  The server 44 includes a communication unit (not shown) and a voice synthesis processing unit 140. When the translated text data is transmitted from the information terminal 20, the server 43 performs speech synthesis processing that causes the speech synthesis processing unit 140 to generate translated speech data of the second language from the translated text data. In addition, when the second specific text information is transmitted from the information terminal 20, the server 43 performs a speech synthesis process that causes the speech synthesis processing unit 140 to generate post-translation speech data in the second language from the second specific text information. Then, the server 43 transmits the generated second specific text information or post-translation voice data of the post-translation text information to the information terminal 20.

  Although FIG. 10 shows an example in which each configuration of the machine translation system 10A is distributed in an information terminal and a plurality of servers, the present invention is not limited to this. For example, each configuration may be distributed to servers smaller than the number of servers illustrated in FIG. 10, or each configuration may be aggregated in one server.

[Operation of Machine Translation System 10A]
Next, the operation of the machine translation system 10 configured as described above will be described. Hereinafter, data exchange between the information terminal 20 and each server will be described with reference to FIG.

  FIG. 12 is a sequence diagram illustrating an example of the operation of the machine translation system 10A according to the modification of the first embodiment.

  As shown in FIG. 12, first, when the voice indicating the utterance of the first language is input from the speaker (S 101), the information terminal 20 sends the pre-translation voice data obtained by converting the voice via the communication unit 30. Then, it is transmitted to the server 41 (S102).

  Next, the server 41 executes speech recognition processing on the received pre-translational speech data, and converts the pre-translational speech data into pre-translation text information in the first language. Then, the server 41 transmits the converted pre-translation text information to the information terminal 20 (S103).

  Next, the information terminal 20 transmits the pre-translation text information received from the server 41 to the server 42 (S104). When the server 42 executes the determination process and determines that the first specific text information is included in the pre-translation text information received from the information terminal 20, the server 42 is associated with the first specific text information. 2 The specific text information is transmitted to the information terminal 20 (S105). If the server 42 determines that the first specific text information is not included in the pre-translation text information received from the information terminal 20, the server 42 transmits the pre-translation text information to the information terminal 20 as not applicable (S105). ). In addition, you may transmit to the information terminal 20 only that there is no corresponding.

  Next, when the information terminal 20 receives the pre-translation text information or not applicable from the server 42, the information terminal 20 transmits the pre-translation text information to the server 43 (S106). Then, the server 43 executes a translation process, and generates post-translation text information of the second language from the pre-translation text information received from the information terminal 20. Then, the server 43 transmits the generated post-translation text information to the information terminal 20 (S107).

  Next, when the translated text information is received from the server 43, the information terminal 20 transmits the translated text information to the server 44 (S108).

  On the other hand, in S105, when the information terminal 20 receives the second specific text information from the server 42, the information terminal 20 skips the translation process and transmits the second specific text information to the server 44 (S108).

  Next, the server 44 executes speech synthesis processing to generate post-translational speech data of the second specific text information or post-translation text information. Then, the server 44 transmits the generated second specific text information or post-translation voice data of the post-translation text information to the information terminal 20 (S109).

  Finally, the information terminal 20 executes voice output of the translated voice data of the second specific text information or the translated text information received from the server 43 (S110).

[Effects]
As described above, according to this modification, the speaker (speaker) only utters the machine translation system 10 including a specific word or specific sentence (first specific text information) in the first language. Thus, it is possible to output a fixed sentence (second specific text information) in the second language that is frequently used, such as an original conversation, a word, and a phrase for each usage. Thereby, the utterance burden of the speaker is reduced.

  The storage unit 163 included in the server 42 stores the second specific text information associated with the first specific text information. However, the present invention is not limited to this. The storage unit 163 may store post-translational voice data of the second specific text information associated with the first specific text information. Hereinafter, the exchange of data between the information terminal 20 and each server in this case will be described.

  FIG. 13 is a sequence diagram illustrating another example of the operation of the machine translation system 10A according to the modification of the first embodiment. Elements similar to those in FIG. 12 are denoted by the same reference numerals, and detailed description thereof is omitted. The sequence diagram shown in FIG. 13 differs from the sequence diagram shown in FIG. 12 in the processing of S105a and S108a, and therefore only the differences will be described below.

  That is, in S105a, when the server 42 executes the determination process and determines that the first specific text information is included in the pre-translation text information received from the information terminal 20, the server 42 corresponds to the first specific text information. The translated speech data of the second specific text information attached is transmitted to the information terminal 20.

  In S108a, when the information terminal 20 receives the translated speech data of the second specific text information from the server 42, the information terminal 20 does not send anything to the server 44, that is, skips the speech synthesis process.

  Therefore, in S110, the information terminal 20 executes voice output of the post-translation voice data of the second specific text information received from the server 42.

(Embodiment 2)
In the machine translation system 10 according to the first embodiment, it has been described that the determination process for determining whether or not the pre-translation text information includes the first specific text information stored in the storage unit 163 is performed. Not limited to that. A determination process for determining whether or not the pre-translation text information and the first specific text information stored in the storage unit 163 completely match may be performed. Hereinafter, this case will be described as a second embodiment, focusing on differences from the first embodiment.

[Configuration of Machine Translation System 10]
The machine translation system 10 according to the second embodiment is different from the machine translation system 10 according to the first embodiment in a translation determination processing unit 16. Since other configurations are the same as those of the machine translation system 10 according to the first embodiment, the description thereof is omitted.

<Translation determination processing unit 16>
The translation determination processing unit 16 according to the second embodiment is different from the translation determination processing unit 16 according to the first embodiment in the operation of the determination unit 162. The other configuration is the same as that of the translation determination processing unit 16 according to the first embodiment, and a description thereof will be omitted.

  In the present embodiment, the determination unit 162 determines whether the pre-translation text information acquired by the acquisition unit 161 matches the first specific text information stored in the storage unit 163. When the determination unit 162 determines that the pre-translation text information and the first specific text information match, the determination unit 162 stores the output unit 164 in the storage unit 163 and associates it with the first specific text information. The second specific text information is output. Since other operations are the same as those described in the first embodiment, description thereof is omitted here.

[Operation of Machine Translation System 10]
A specific aspect of the operation of the machine translation system 10 according to the second embodiment configured as described above will be described.

  FIG. 14 is a flowchart showing a specific mode of operation of the machine translation system in the second embodiment. The same elements as those in FIG. 7 are denoted by the same reference numerals, and detailed description thereof is omitted. That is, the processing in S11, S12, and S14 to S17 shown in FIG. 14 is the same as that described in the first embodiment, and thus the description thereof is omitted. Hereinafter, a determination process including S23 different from that of the first embodiment will be described.

  In S23, the translation determination processing unit 16 performs a determination process. That is, the translation determination processing unit 16 determines whether or not the acquired pre-translation text information matches a specific word or a specific sentence registered in advance in the storage unit 163 (S23). In S23, if the pre-translation text information matches the specific word or specific sentence registered in advance in the storage unit 163 (Y in S23), it is stored in the storage unit 163 and corresponds to the first specific text information. The attached second specific text information is extracted (S14).

  Here, it demonstrates more concretely using FIG.

  For example, it is assumed that the speaker utters “I will explain the translator” in Japanese to the machine translation system 10 according to the second embodiment. As shown in FIG. 8, the storage unit 163 stores the specific sentence “translation machine description” as the first specific text information, but the utterance content of the speaker is not the word “translation machine description”. Also includes the word “I do”. In this case, in S23, the machine translation system 10 in Embodiment 2 determines that the utterance content of the speaker and the first specific text information do not completely match (N in S23), and proceeds to S15. The machine translation system 10 according to the second embodiment executes a translation process in S15 and outputs post-translation text information of the second language “I explain this translator machine.” To the speech synthesizer 14 as a result.

  On the other hand, it is assumed that the speaker utters “translation machine explanation” in Japanese to the machine translation system 10 according to the second embodiment. As shown in FIG. 8, the storage unit 163 stores a specific sentence “description of the translator” as first specific text information. In this case, in S23, the machine translation system 10 in the second embodiment determines that the utterance content of the speaker and the first specific text information completely match (Y in S23), and proceeds to S14.

[Effects]
As described above, the machine translation system 10 according to the second embodiment outputs the second specific text information when the pre-translation text information matches the first specific text information. That is, when the speaker (speaker) utters only the first specific text information, the machine translation system 10 outputs the second specific text information as a translation result, and the speaker (speaker) outputs the first specific text. When an utterance including a word other than information is made, a translated sentence obtained by translating the uttered content as it is is output.

  As a result, a speaker using the machine translation system 10 can use the specific word or specific sentence indicated by the first specific text information in the conversation.

  In other words, for the other party (listener), the speaker uses a fixed sentence that has been registered in advance and is a frequently used fixed sentence for a registered specific sentence such as “Explanation of translator”. Suppose you want to. In this case, the speaker can cause the machine translation system 10 to output a fixed sentence (second specific text information) by speaking only the registered specific word (first specific text information).

  On the other hand, when the other party cannot understand the contents of the pre-registered fixed sentence (second specific text information), the speaker uses a different utterance expression from the pre-registered fixed sentence to the listener. There are times when I want to explain. At this time, if the machine translation system 10 includes a registered specific sentence (first specific text information) such as “Explanation of translator” in the utterance content (pre-translation text information), the specific sentence (first specific text information) When a fixed sentence corresponding to (specific text information) is output, the speaker is restricted from including the first specific text information in the utterance content. Therefore, in the second embodiment, a determination condition is used as to whether or not the pre-translation text information matches the first specific text information. Thus, the speaker can freely perform the utterance content including the registered specific sentence (first specific text information) such as “Explanation of the translator” and cause the machine translation system 10 to perform the translation process. As described above, the speaker can flexibly select an explanation method by outputting a fixed sentence to the listener and explaining it by using his / her own words.

  Also, for example, in the case of explaining a translator, if the listener has previously used a translator, the speaker will ask the listener, “Is it unnecessary to explain the translator?” It is assumed that At this time, if the machine translation system 10 includes a registered specific sentence such as “Explanation of translator” in the utterance content, the confirmation message is output when a fixed sentence corresponding to the specific sentence is output. Can not. On the other hand, in the present embodiment, the machine translation system 10 outputs a translated sentence in the second language in which the utterance content “Is the explanation of the translator unnecessary?” Is output, and the confirmation can be performed by the listener. . In this way, according to the machine translation system 10 of the present embodiment, the speaker can also confirm the presence / absence of the explanation of the translator in a natural conversation, and is fixed for customers who do not need explanation of the translator. It is also possible to prevent the situation where the explanation of the translator, which is a sentence, is output.

  Furthermore, according to the machine translation system 10 of the present embodiment, the second specific text information is output only when the pre-translation text information completely matches the specific word or the specific sentence that is the first specific text information. . Therefore, the speaker can also eliminate the unnecessary concern that the speaker is aware of the content of the utterance so that the specific word or the specific sentence is not included in the normal conversation.

  The machine translation system 10 may include a display as in the first embodiment. In this case, the machine translation system 10 performs a determination process, and outputs a standard English sentence stored in the storage unit 163 and corresponding to the specific sentence to a listener whose native language is English. A fixed sentence translation (third specific text information) corresponding to the fixed sentence is displayed to a speaker whose native language is Japanese. Accordingly, when outputting the voice of the fixed sentence (second specific text information) of the second language that is different from the specific sentence of the first language spoken by the speaker but is associated, what kind of fixed sentence is to be output It is possible to inform the speaker whether the sound is output. Also, even when the speaker does not memorize the contents of the fixed sentence, the contents of the fixed sentence in the second language can be confirmed on the spot with the fixed sentence translated sentence (the fixed sentence in the first language). The conversation can be made smoothly.

(Modification)
In the above, Embodiment 1 and Embodiment 2 have been described as different embodiments, but they can also be implemented in combination. An example of this case will be described with reference to FIG.

  In FIG. 8, a “type” item is provided. For example, in a specific sentence in which the value of the “type” is set to “1”, when the specific sentence (first specific text information) is included in the utterance content (pre-translation text information) of the speaker, The machine translation system 10 may output a fixed sentence (second specific text information) corresponding to the specific sentence. On the other hand, in the specific sentence in which the attribute indicated by “type” is set to “2”, the utterance content (pre-translation text information) of the speaker and the specific sentence (first specific text information) completely match. Only the fixed sentence (second specific text information) corresponding to the specific sentence may be output to the machine translation system 10.

  As a result, specific sentences that are rarely used are replaced with pre-registered second specific text information if they are included anywhere in the speaker's utterance, and specific words and specific sentences that are frequently used are completely Only when they match, the determination process can be performed with individual rules such as replacement with the second specific text information registered in advance. That is, the determination process can be performed for each specific sentence using individual rules. This means that easy-to-use rules can be individually set according to the usage frequency of the specific word to be registered and the specific sentence itself in normal conversation. That is, the convenience of the speaker (user) can be further improved.

(Embodiment 3)
In the third embodiment, an example of a determination process different from the determination process described in the first and second embodiments will be described.

[Configuration of Machine Translation System 10]
The machine translation system 10 according to the third embodiment is different from the machine translation system 10 according to the first embodiment in a translation determination processing unit 16. Since other configurations are the same as those of the machine translation system 10 according to the first embodiment, the description thereof is omitted.

<Translation determination processing unit 16>
The translation determination processing unit 16 according to the second embodiment is different from the one stored in the storage unit 163 and the operation of the determination unit 162 as compared to the translation determination processing unit 16 according to the first embodiment. The other configuration is the same as that of the translation determination processing unit 16 according to the first embodiment, and a description thereof will be omitted.

  In the present embodiment, storage unit 163 is expressed in first specific text information indicating a specific word or specific sentence expressed in the first language, and in a second language that is a language different from the first language prepared in advance. Second specific text information indicating a fixed phrase that is a word or sentence, the specific word or specific sentence and second specific text information having no translation equivalence are stored in association with each other.

  Furthermore, in each of the second specific text information, the storage unit 163 includes two or more first specific text information and order information indicating the order in which each of the two or more first specific text information should appear in the sentence. It is stored in association.

  The determination unit 162 includes the two or more first specific text information stored in the storage unit 163 in the pre-translation text information, and the two or more first specific text information indicates the order indicated by the order information. It is determined whether it is included.

  Here, the determination unit 162 includes the two or more first specific text information stored in the storage unit 163 in the pre-translation text information, and the two or more first specific text information is the order information. It is assumed that they are included in the order indicated by In this case, the determination unit 162 causes the output unit 164 to output two or more first specific text information and second specific text information corresponding to the order information.

  Other configurations are the same as those described in the first embodiment, and a description thereof is omitted here.

  As described above, the translation determination processing unit 16 according to the third embodiment uses a specific word or specific sentence (first specific text information) registered in advance in the utterance contents (pre-translation text information) of the speaker. It is determined whether or not they are included in the registered utterance order. And, when the specific word or the like is included in the utterance contents of the speaker in the order of the utterances registered in advance, it is registered in advance corresponding to the specific word or the like without translation processing. A fixed sentence (second specific text information) is output.

[Operation of Machine Translation System 10]
A specific aspect of the operation of the machine translation system 10 according to the third embodiment configured as described above will be described with reference to FIGS. 15 and 16.

  FIG. 15 is a flowchart showing a specific mode of operation of machine translation system 10 in the third embodiment. The same elements as those in FIG. 7 are denoted by the same reference numerals, and detailed description thereof is omitted. That is, since the processes in S11, S12, and S14 to S17 shown in FIG. 15 are as described in the first embodiment, the determination process including S33 and S34 different from the first embodiment will be described below.

  In S33 and S34, the translation determination processing unit 16 performs determination processing. That is, the translation determination processing unit 16 determines whether or not the acquired pre-translation text information includes a plurality of specific words or a plurality of specific sentences registered in advance in the storage unit 163 (S33).

  In S33, when the plurality of specific words or the plurality of specific sentences are included in the pre-translation text information (Y in S33), the translation determination processing unit 16 stores the plurality of specific words or the plurality of specific sentences. Then, it is determined whether or not they are included in the order indicated by the order information registered in advance in the storage unit 163 (S34). That is, in S34, the translation determination processing unit 16 determines in what order the plurality of specific words or the plurality of specific sentences are included in the pre-translation text information. More specifically, in S34, the translation determination processing unit 16 determines whether the order of the plurality of specific words or the plurality of specific sentences matches the order indicated by the order information stored in the storage unit 163 in advance. Determine whether.

  In S34, the translation determination processing unit 16 stores the plurality of specific words or the plurality of specific sentences in the order indicated by the order information (Y in S34) and is stored in the storage unit 163. And the 2nd specific text information matched with the said some specific word or the said some specific sentence is extracted (S14). When it is determined in S33 that the plurality of specific words or the plurality of specific sentences are not included in the pre-translation text information, or in S34, the plurality of specific words or the plurality of specific words are included in the pre-translation text information. If it is determined that the sentences are not included in the order indicated by the order information, the process proceeds to S15.

  FIG. 16 is a diagram illustrating an example of fixed phrases corresponding to specific words and utterance orders stored in association with each other in the storage unit according to the third embodiment.

  In FIG. 16, a plurality of Japanese specific words are shown as an example of the first specific text information, and the utterance order is shown as an example of order information indicating the order in which the plurality of specific words should appear in the sentence. ing. FIG. 16 shows an English fixed sentence as an example of the second specific text information. Further, FIG. 16 shows a fixed sentence translated sentence obtained by translating an English fixed sentence into Japanese as an example of the third specific text information.

  More specifically, in FIG. 16, for example, the specific word “guidance” is associated with the order information “(1)”, and the specific word “Tokyo Station” is in the order “(2)”. It is shown that the information is stored in the storage unit 163 in association with the information.

  For example, when a speaker speaks “Guide, Tokyo Station” in Japanese to the machine translation system 10, the machine translation system 10 uses speech recognition processing to convert the speech data spoken by the speaker into Japanese pre-translation text. Convert to information. Next, in the machine translation system 10, determination processing is performed by the translation determination processing unit 16, and “guidance” and “Tokyo Station” which are a plurality of specific words stored in the storage unit 163 are included in the pre-translation text information. Determine that it is included. Further, the machine translation system 10 determines whether or not the plurality of specific words are included in the pre-translation text information in the order indicated by the order information associated with each specific word. That is, in the storage unit 163, the order indicated by the order information associated with the specific word “guidance” is “(1)”, and the order information associated with the specific word “Tokyo Station” indicates the order. The order is “(2)”. Then, the machine translation system 10 determines that the specific words “guidance” and “Tokyo Station” are included in this order in the pre-translation text information. In other words, the machine translation system 10 determines that the order indicated by the order information matches the determined order. For this reason, the machine translation system 10 uses “We will guide you to Tokyo Station” which is a fixed sentence (second text information) corresponding to “Guide” and “Tokyo Station” (first text information) that are a plurality of specific words. First, please turn left out from this building and go as straight 100m.And then, you will find it on the right-hand side. '' Output audio of English fixed phrases.

  On the other hand, in order to confirm whether or not the listener needs guidance to Tokyo Station, the speaker speaks in Japanese to the machine translation system 10 “Would you like to give guidance to Tokyo Station?” Suppose. In this case, the machine translation system 10 converts the voice data of the voice uttered by the speaker into Japanese pre-translation text information by voice recognition processing. Next, in the machine translation system 10, the translation determination processing unit 16 performs a determination process, and the pre-translation text information includes a plurality of specific words “Tokyo Station” and “Guidance” stored in the storage unit 163. It is determined that Further, the machine translation system 10 determines whether or not the plurality of specific words are included in the pre-translation text information in the order indicated by the order information associated with each specific word. That is, in the storage unit 163, the order indicated by the order information associated with the specific word “Tokyo Station” is “(1)”, and the order information associated with the specific word “guidance” indicates the order. The order is “(2)”. Then, the machine translation system 10 determines that the specific words “guidance” and “Tokyo Station” are included in this order in the pre-translation text information. That is, the machine translation system 10 determines that the order indicated by the order information does not match the determined order. For this reason, the machine translation system 10 performs a translation process for translating a Japanese sentence “Would you like to give guidance to Tokyo?” Spoken by a speaker into English, for example, “Would you wish a guide to Tokyo Station” ? "Is translated into English sentences, voice synthesis processing and voice output processing are executed, and the voices of the above English sentences are output.

  The same applies to the case where the speaker utters the machine translation system 10 using a plurality of specific words “explanation” and “checkout”, and the description is omitted here.

  In addition, the machine translation system 10 may include a display as in the first embodiment. In this case, the machine translation system 10 performs a determination process, and outputs a standard English sentence stored in the storage unit 163 and corresponding to the specific sentence to a listener whose native language is English. A fixed sentence translation (third specific text information) corresponding to the fixed sentence is displayed to a speaker whose native language is Japanese. Accordingly, when outputting the voice of the fixed sentence (second specific text information) of the second language that is different from the specific sentence of the first language spoken by the speaker but is associated, what kind of fixed sentence is to be output It is possible to inform the speaker whether the sound is output. Also, even when the speaker does not memorize the contents of the fixed sentence, the contents of the fixed sentence in the second language can be confirmed on the spot with the fixed sentence translated sentence (the fixed sentence in the first language). The conversation can be made smoothly.

[Effects]
As described above, according to the third embodiment, whether the speaker (speaker) uses the second specific text information, which is a fixed sentence frequently used in advance, registered in the machine translation system 10. Whether or not to speak can be easily switched depending on whether or not to utter in consideration of the utterance order of a plurality of specific words or a plurality of specific sentences.

  Also, for example, when the first language is Japanese, as shown in FIG. 16, in a normal Japanese conversation, the verb is a specific word representing “verb” by making use of the grammatical features included at the end of the sentence. Or, order information “(1)” is set for a specific sentence, and order information “(2)” is set for a specific word or a specific sentence representing “subject”, “object”, etc. do it. Thereby, it is possible to prevent the second specific text information from being output when the specific word or the specific sentence is included in the utterance sentence, not for the purpose of outputting the second specific text information. Moreover, in the utterer, whether to output the second specific text information as the translation result for the utterance sentence including the first specific text information or whether to output the translated sentence translated as it is, It can be selected as appropriate according to the manner of utterance.

  That is, according to the machine translation system of the third embodiment, the second specific text information is intentionally spoken differently from the normal grammar without affecting the translation process of the speaker's normal conversation. The effect of output can be easily obtained. Further, according to the machine translation system in the third embodiment, even a speaker who does not know the output condition of the second specific text information stored in the storage unit 163 is as long as he / she speaks in accordance with normal grammar. The second specific text information can be prevented from being output unintentionally.

  Note that “verb” and “other parts of speech” pairs may be registered in the storage unit 163 as a plurality of specific words or a plurality of specific sentences. For example, a specific word “guidance” is set in “verb”, and “(1)” is set in association with the specific word as order information. On the other hand, a plurality of specific words of “other parts of speech” to be paired can be set. For example, “Osaka Station”, “Tokyo Sky Tree”, “Toilet”, etc. At this time, as the order information for “other parts of speech”, “(2)” may be set in association with each other. In this way, by storing a plurality of specific words or pairs of specific sentences in the storage unit 163 in association with 1 to n, the second specific text information can be added mechanically, and the storage unit 163 Easy maintenance, such as adding, updating, and avoiding duplicate entries.

  As described above, the machine translation method and the machine translation system according to one or more aspects of the present disclosure have been described based on the embodiments, but the present disclosure is not limited to the embodiments. Unless it deviates from the gist of the present disclosure, one or more of the present disclosure may be applied to various modifications conceived by those skilled in the art in the present embodiment, or forms configured by combining components in different embodiments. It may be included within the scope of the embodiments.

  For example, in the machine translation method and the machine translation system of the present disclosure, as described with reference to FIG. 10, the speech recognition process, the translation determination process, the translation process, and the speech synthesis process are executed on different independent servers. May be. Of course, some of the plurality of processes may be executed by the same server, or all processes may be executed by the same server. In any form, the same effect can be obtained.

  The plurality of processes do not necessarily need to be executed on the server via a communication means such as a network, and some of them are executed in the information terminal, that is, information connected via an internal bus. You may perform in the function structure in a terminal. Some of them may be executed by a peripheral device directly connected to the information terminal.

  In addition, in the machine translation method and the machine translation system of the present disclosure, when the spoken content is translated into multiple languages and transmitted to the other party, as described above, not only is transmitted to the other party by multilingual voice output. Alternatively, the text may be displayed on the display to the other party. This is because the same effect can be obtained.

  In addition, the technique demonstrated in the said aspect can be implement | achieved in the following types of cloud services, for example. However, the types of cloud services in which the technology described in the above aspect is realized are not limited to these.

(Service type 1: In-house data center type cloud service)
FIG. 17 is a diagram showing an overview of services provided by the information management system in service type 1 (in-house data center type cloud service). In this type, the service provider 11200 acquires information from the group 11000 and provides a service to the user. In this type, the service provider 11200 has a function of a data center operating company. That is, the service provider 11200 has a cloud server 11110 that manages big data. Therefore, there is no data center operating company.

  In this type, the service provider 11200 operates and manages a data center (cloud server) 12030. The service provider 11200 manages an operating system (OS) 12020 and an application 12010. The service provider 11200 provides a service using the OS 12020 and the application 12010 managed by the service provider 11200 (arrow 12040).

(Service type 2: Cloud service using IaaS)
FIG. 18 is a diagram illustrating an overall image of services provided by the information management system in service type 2 (IaaS-based cloud service). Here, IaaS is an abbreviation for infrastructure as a service, and is a cloud service provision model that provides a base for constructing and operating a computer system as a service via the Internet.

  In this type, the data center operating company 11100 operates and manages the data center (cloud server) 12030. The service provider 11200 manages the OS 12020 and the application 12010. The service provider 11200 provides a service using the OS 12020 and the application 12010 managed by the service provider 11200 (arrow 12040).

(Service type 3: Cloud service using PaaS)
FIG. 19 is a diagram illustrating an overall image of services provided by the information management system in service type 3 (PaaS use type cloud service). Here, PaaS is an abbreviation for Platform as a Service, and is a cloud service provision model that provides a platform serving as a foundation for constructing and operating software as a service via the Internet.

  In this type, the data center operating company 11100 manages the OS 12020 and operates and manages the data center (cloud server) 12030. The service provider 11200 manages the application 12010. The service provider 11200 provides a service using the OS 12020 managed by the data center operating company 11100 and the application 12010 managed by the service provider 11200 (arrow 12040).

(Service type 4: Cloud service using SaaS)
FIG. 20 is a diagram illustrating an overview of services provided by the information management system in service type 4 (SaaS-based cloud service). Here, SaaS is an abbreviation for software as a service. The SaaS-based cloud service is, for example, an application provided by a platform provider who owns a data center (cloud server), or a user such as a company or individual who does not have a data center (cloud server) on the Internet. This is a cloud service provision model that has functions that can be used via other networks.

  In this type, the data center operating company 11100 manages the application 12010, manages the OS 12020, and operates and manages the data center (cloud server) 12030. Further, the service provider 11200 provides a service using the OS 12020 and the application 12010 managed by the data center operating company 11100 (arrow 12040).

  As described above, in any cloud service type, the service provider 11200 provides a service. In addition, for example, the service provider or the data center operating company may develop an OS, an application, or a big data database, or may be outsourced to a third party.

  The present disclosure can be used for a machine translation method and a machine translation system, and in particular, can be used for a machine translation system and a machine translation method in a machine translation system that can be easily used, such as a PC app, a web app, and a smartphone app. it can.

10, 10A, 90 Machine translation system 11, 91 Speech input unit 12, 92 Speech recognition unit 13, 93 Translation unit 14, 94 Speech synthesis unit 15, 95 Speech output unit 16, 16A Translation judgment processing unit 17, 23 Display 20 information Terminal 21 Communication unit 22, 163 Storage unit 30 Communication means 41, 42, 43, 44 Server 50 Speaker 51 Listener 110 Speech recognition processing unit 130 Translation processing unit 140 Speech synthesis processing unit 161 Acquisition unit 162, 162A Determination unit 164 Output unit

Claims (12)

A machine translation method in a machine translation system,
In the storage unit, the first specific text information indicating the specific word or specific sentence expressed in the first language, and the word or sentence expressed in the second language which is a language different from the first language prepared in advance. Corresponds to at least one of the second specific text information indicating the fixed sentence, the second specific text information having no translation equivalence with the specific word or the specific sentence, and the second audio data of the second specific text information Attached and remembered,
An acquisition step of acquiring pre-translation text information generated by converting first speech data representing input speech uttered in the first language into text information;
A determination step of determining whether or not the pre-translation text information acquired in the acquisition step includes the first specific text information stored in the storage unit;
In the determination step, when it is determined that the pre-translation text information includes the first specific text information, the text information is stored in the storage unit and associated with the first specific text information. An output step of outputting at least one of the second specific text information and the second audio data of the second specific text information.
Machine translation method. In the determination step, it is determined whether or not the pre-translation text information matches the first specific text information stored in the storage unit;
In the output step, when it is determined in the determination step that the pre-translation text information and the first specific text information match, the output step is stored in the storage unit and is associated with the first specific text information. Outputting at least one of the second specific text information and the second audio data of the second specific text information,
The machine translation method according to claim 1. In the storage unit, the second specific text information includes two or more first specific text information, and order information indicating an order in which each of the two or more first specific text information should appear in a sentence; Are stored in association with each other,
In the determination step, the pre-translation text information acquired in the acquisition step includes the two or more first specific text information stored in the storage unit, and the two or more first specific text information Are included in the order indicated by the order information,
In the output step, in the determination step, the pre-translation text information includes the two or more first specific text information stored in the storage unit, and the two or more first specific text information includes When it is determined that they are included in the order indicated by the order information, the second specific text information corresponding to the two or more first specific text information and the order information and the second specific text information corresponding to the second specific text information. 2 Output at least one of the audio data,
The machine translation method according to claim 1. In the storage unit, one second specific text information includes one or more first specific text information that are different one or more first specific text information, each including the same specific word and indicating a different specific sentence. Stored in association,
The machine translation method according to claim 1 or 2. In the output step, when it is determined in the determination step that the first specific text information is not included in the pre-translation text information, the post-translation text information translated into the second language Text information is output,
The machine translation method according to claim 1. In the output step, when it is determined in the determination step that the pre-translation text information includes the first specific text information stored in the storage unit, the post-translation text information is not output.
The machine translation method according to claim 5. In the storage unit, third specific text information obtained by translating the second specific text information into the first language is used as at least the second specific text among the first specific text information and the second specific text information. Stored in association with information,
In the output step, when outputting at least one of the second specific text information and the second audio data of the second specific text information, the third specific text information is further output.
The machine translation method according to claim 1. The third specific text information output in the output step is displayed on a display.
The machine translation method according to claim 7. The machine translation system is connected to an information terminal having a display by a predetermined communication means,
In the output step, at least one of the second specific text information and the second audio data of the second specific text information is output to the information terminal by the predetermined communication means.
The machine translation method according to claim 1. In the information terminal, when the second specific text information is output in the output step, speech synthesis processing is performed on the second specific text information to generate second voice data, and the generated second voice data indicates Output audio,
The machine translation method according to claim 9. The machine translation method is used in a predetermined situation between a person who speaks the first language and a person who speaks the second language.
The machine translation method according to claim 1. First specific text information indicating a specific word or specific sentence expressed in the first language, and a fixed sentence that is a word or sentence expressed in a second language that is different from the first language prepared in advance. A storage unit that stores second specific text information that is associated with at least one of the second specific text information that is not translationally equivalent to the specific word or the specific sentence and the second audio data of the second specific text information. When,
An acquisition unit for acquiring pre-translation text information generated by converting first voice data representing input voice uttered in the first language into text information;
A determination unit that determines whether or not the pre-translation text information acquired in the acquisition unit includes the first specific text information stored in the storage unit;
When the determination unit determines that the first specific text information is included in the pre-translation text information, the storage unit stores the first specific text information and associates the first specific text information with the first specific text information. An output unit that outputs at least one of the second specific text information and the second audio data of the second specific text information.
Machine translation system.

Images