KR20200095040A - Server and method for multilingual chat - Google Patents

Abstract

Disclosed are a server for translation and a method thereof. According to one embodiment, the server may obtain at least one relay language and at least one optimum translation engine from a database based on the input language and the target language and generate the optimum translation path specified by the at least one relay language and the at least one optimum translation engine. According to one embodiment, the server can generate the optimum translation path for translation between the input language and the target language based on translation information between the input language and the target language translation information between the input language and a plurality of languages, and information between the plurality of languages and the target language.

Description

Server and translation method for multilingual chat {SERVER　AND METHOD FOR MULTILINGUAL CHAT}

The following embodiments relate to a translation technique for multilingual chat.

With the development of internet technology and mobile technology, a chat service between a number of smart terminals is active. Existing chat systems are systems developed for users who use the same language, and chat with each other is possible only when using the same language. Real-time chat is difficult for users who speak different languages. Therefore, there is an inconvenience of having to chat in either language of the parties or chat in English that can be recognized as a common language. Alternatively, users write a message in their native language in a chat window, and use a separate translation program to translate the message into the other user's language and then chat. According to the conventional chat system, there is a problem in that it is difficult to provide a seamless chat service between users using different languages.

In translation technology, as Internet technology becomes more common, attempts to expand or integrate the platform of network-based services are increasing.As part of the platform expansion, improvement in translation technology performance to provide smooth communication between different languages is increasing. Is required. Recently, since machine learning-based translation techniques are widely adopted, the amount of secured learning data becomes an important factor that makes a difference in the performance of the translation engine. However, since the amount of learning data is different for each language, the difference in the amount of learning data may lead to a performance gap of the translation engine between languages. If the performance gap of the translation engine increases, the development of translation technology for languages in regions with poor IT (Information Technology) environments may be delayed. Therefore, it is required to develop a translation technology that considers the performance of translation between languages.

Embodiments are intended to provide multilingual voice chat.

Embodiments are intended to improve the translation accuracy of the translation technology.

The embodiments seek to bridge the gap in translation services between languages.

The embodiments are intended to provide an optimal translation path in consideration of translation information for each language of a translation engine.

Embodiments are intended to provide a location-based speech recognition translation and speech synthesis technology.

In a multilingual chat according to an embodiment, a method of operating a server includes: receiving information of participants in multilingual chat; Receiving input language text from the chat sender; Determining a target language for each of the chat recipients based on the information of the multilingual chat participants; Obtaining translation information for each language of the translation engines; Obtaining first translation information for the input language and the target language of the translation engines based on the translation information for each language of the translation engines; Obtaining second translation information for the input language and a plurality of languages of the translation engines based on the translation information for each language of the translation engines; Acquiring the plurality of languages of the translation engines and third translation information for the target language based on the translation information for each language of the translation engines; And generating an optimal translation path of the input language text based on the first translation information, the second translation information, and the third translation information.

Generating the optimal translation path may include determining at least one optimal translation engine corresponding to the optimal translation path.

The translation information for each language of the translation engines may include at least one of translation information for each language of the translation engines, a time required for translation corresponding to the translation information for each language of the translation engines, and a regulation for each language of the translation engines. .

The generating of the optimal translation path may include selecting at least one relay language from the plurality of languages based on the second translation information and the third translation information; Comparing the first translation information with fourth translation information of a translation path that is a sequence of the input language, the at least one relay language, and the target language; And determining one of a translation path corresponding to the first translation information and a translation path corresponding to the fourth translation information as the optimum translation path based on the comparison result.

In the multilingual chat according to an embodiment, the method of operating the server further includes receiving location information of the chat sender, and translation information of the translation engines for each language may include the location information of the chat sender. have.

The information of the participants may include the language used by the participants.

The translation information between the first language and the second language includes an amount of translation data held between the first language and the second language, a word order match rate between the first language and the second language, and the capacity of the characters in the first language. Is updated based on at least one of a match of the capacity of a character of the second language and a feedback corresponding to a translation between the first language and the second language, and the first language and the second language are the input A language, the target language, and the at least one relay language may be included.

In a multilingual chat according to an embodiment, a method of operating a server may include receiving a target language text translated into a target language of each of the chat recipients based on the optimal translation path; And transmitting each of the target language texts to the chat recipients.

The input language text may include text obtained by converting the voice of the input language.

In one embodiment, a method of operating an input terminal in a multilingual chat includes: receiving an input language voice from a chat sender; Obtaining conversion information for each language of the conversion engines; Converting the input language speech into input language text based on the conversion information for each language of the conversion engines; Displaying the input language text; And transmitting information of multilingual chat participants, the input language voice, and the input language text to a server, wherein the server

The target language of each chat recipient is determined based on the information of the multilingual chat participants, and an optimal translation path of the input language text is generated based on translation information of each language of the translation engines.

In one embodiment, a method of operating an input terminal in a multilingual chat includes: acquiring location information of the chat sender; Transmitting the location information of the chat sender to a server; And converting the input language voice into the input language text based on the location information.

Converting the input language voice into the input language text may include converting the input language voice into the input language text based on an artificial neural network.

In a multilingual chat according to an embodiment, a method of operating a target terminal includes: receiving a target language text from a server; Displaying the target language text; Obtaining synthesis information for each language of the synthesis engines; Synthesizing the target language text into a target language speech based on the synthesis information for each language of the synthesis engines; And reproducing the target language voice, wherein the server determines a target language for each of the chat recipients based on the information of the multilingual chat participants, and the input language text based on translation information for each language of the translation engines. Create an optimal translation path.

In the multilingual chat according to an embodiment, the operation method of the target terminal may further include not receiving the target language text, but receiving only the target language voice when the input language and the target language are the same.

The server according to an embodiment includes a database for recording translation information of each language of translation engines and information of multilingual chat participants; And a processor for generating an optimal translation path between the input language and the target language, wherein the processor provides information on the input language and the target language of the translation engines based on the translation information for each language of the translation engines. 1 Acquire translation information, obtain second translation information for the input language and a plurality of languages of the translation engines based on the translation information for each language of the translation engines, and Based on the translation information, the plurality of languages of the translation engines and third translation information for the target language are obtained, and the first translation information, the second translation information, and the third translation information Based on, at least one optimal translation engine is determined, and based on the at least one optimal translation engine, an optimal translation path of the input language text is generated.

The processor may determine at least one optimal translation engine corresponding to the optimal translation path.

The translation information for each language of the translation engines may include at least one of translation information for each language of the translation engines, a time required for translation corresponding to the translation information for each language of the translation engines, and a regulation for each language of the translation engines. .

The processor selects at least one relay language from the plurality of languages based on the second translation information and the third translation information, and selects the input language, the at least one relay language, and the target language. The fourth translation information of the translation path which is a sequence is compared with the first translation information, and based on the comparison result, any one of a translation path corresponding to the first translation information and a translation path corresponding to the fourth translation information May be determined as the optimal translation path.

The processor may receive location information of the chat sender, and the translation information of the translation engines for each language may include the location information of the chat sender.

The information of the participants may include the language used by the participants.

Embodiments may provide multilingual voice chat.

Embodiments can improve the translation accuracy of the translation technology.

Embodiments can bridge the gap in translation services between languages.

Embodiments may provide an optimal translation path in consideration of translation information for each language of a translation engine.

Embodiments may provide location-based speech recognition translation and speech synthesis technology.

1 is a diagram for describing a multilingual chat system according to an exemplary embodiment.
2 is a diagram illustrating a server for translation according to an embodiment.
3 is a flowchart illustrating a method of generating an optimal translation path according to an exemplary embodiment.
4 is a diagram illustrating translation information for each language of translation engines according to an exemplary embodiment.
5 is a diagram illustrating a process of generating an optimal translation path according to an embodiment.
6 is an exemplary diagram of a configuration of a server according to an embodiment.
7 is a flowchart illustrating a method of operating an input language terminal according to an embodiment.
8 is a flowchart illustrating a method of operating a target language terminal according to an embodiment.

Specific structural or functional descriptions of embodiments according to the concept of the present invention disclosed in the present specification are exemplified only for the purpose of describing embodiments according to the concept of the present invention, and embodiments according to the concept of the present invention They may be implemented in various forms and are not limited to the embodiments described herein.

Embodiments according to the concept of the present invention can be applied to various changes and can have various forms, so the embodiments will be illustrated in the drawings and described in detail herein. However, this is not intended to limit the embodiments according to the concept of the present invention to specific disclosed forms, and includes changes, equivalents, or substitutes included in the spirit and scope of the present invention.

Terms such as first or second may be used to describe various components, but the components should not be limited by the terms. The above terms are only for the purpose of distinguishing one component from other components, for example, without departing from the scope of rights according to the concept of the present invention, the first component may be referred to as the second component, Similarly, the second component may also be referred to as the first component.

When a component is referred to as being “connected” or “connected” to another component, it is understood that it may be directly connected or connected to the other component, but other components may exist in the middle. Should be. On the other hand, when a component is referred to as being “directly connected” or “directly connected” to another component, it should be understood that there is no other component in the middle. Expressions that describe the relationship between components, for example, “between” and “just between” or “directly adjacent to” should be interpreted as well.

The terms used in the present specification are only used to describe specific embodiments and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In the present specification, terms such as “comprise” or “have” are intended to designate that a set feature, number, step, action, component, part, or combination thereof exists, but one or more other features or numbers, It is to be understood that the possibility of addition or presence of steps, actions, components, parts, or combinations thereof is not preliminarily excluded.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person skilled in the art to which the present invention pertains. Terms as defined in a commonly used dictionary should be interpreted as having a meaning consistent with the meaning in the context of the related technology, and should not be interpreted as an ideal or excessively formal meaning unless explicitly defined in the present specification. Does not.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. The same reference numerals in each drawing denote the same members.

1 is a diagram for describing a multilingual chat system according to an exemplary embodiment.

Referring to FIG. 1, a multilingual chat system 100 according to an embodiment includes a first user terminal 110, a server 130, a translation engine 150, a speech synthesis engine 170, and a second user terminal 190. ) Can be included.

According to the multilingual chat system 100, chat participants may be two or more, and chat texts of different languages inputted to each of the chat participants' terminals are translated in real time to provide the original text and the translated text to each user terminal together. I can. In addition, it is possible to recognize the chat text from the voice input of different languages inputted from each of the chat participants' terminals, translate the text in real time, and provide the translated text to the chat counterpart user terminal.

The first user terminal 110, the server 130, the translation engine 150, the speech synthesis engine 170, and the second user terminal 190 may be interconnected through a network. A network refers to a connection structure in which information exchange is possible between each node such as a terminal and a server, and examples of such a network include a 3rd Generation Partnership Project (3GPP) network, a Long Term Evolution (LTE) network, a 5G network, and WIMAX. (World Interoperability for Microwave Access) Network, Internet, LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network), PAN (Personal Area Network), Bluetooth Network, Satellite A broadcasting network, an analog broadcasting network, and a Digital Multimedia Broadcasting (DMB) network are included, but are not limited thereto.

The first user terminal 110 and the second user terminal 190 may chat with each other via the server 130. In FIG. 1, the server 130 is shown to provide a translation chat service to the first user terminal 110 and the second user terminal 190, but the second user terminal 110 is connected to the server 110 to receive the translation chat service. The number of user terminals is not limited to two. A user of the first user terminal 110 may be a chat sender, and a user of the second user terminal 190 may be a chat recipient.

According to various embodiments of the present invention, each of the first user terminal 110 and the second user terminal 190 may be various types of devices. For example, each of the first user terminal 110 and the second user terminal 190 may be a TV device, a computer, or a portable terminal capable of accessing a remote server or device through a network. Here, an example of a TV device includes a smart TV, an IPTV set-top box, etc., an example of a computer includes a notebook, a desktop, and a laptop, and an example of a portable terminal has portability and mobility. As a guaranteed wireless communication device, PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication) -2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (Wireless Broadband Internet) terminals, all kinds of handhelds such as smartphones, tablet PCs, etc. )-Based wireless communication device may be included.

An application for providing a translation chat service to a user may be installed and driven in the first user terminal 110 and the second user terminal 190. For example, the application may provide a user interface for a user to select a chat partner, select a language to be used during a chat, input text in a selected language, or output text translated in a selected language.

The first user terminal 110 may receive an input language voice (111). The input language is a language of the text to be translated, and may be a language for each country such as Korean, Chinese, English, and Spanish. Here, the text includes characters, words, and sentences expressed in the input language. For example, when the input language is Korean, the text of the input language may be expressed by Korean characters, words, and sentences, such as "ㄱ", "school", and "hello".

In addition, the input language voice is a message that the user of the first user terminal 110 wants to transmit to the chatting partner, and may be a voice input by the user through the first user terminal 110.

The first user terminal 110 may receive an input for selecting an input language prior to receiving the input language voice. For example, the user may select an input language to be used to input an input language voice through a touch interface and an application provided by the first user terminal 110.

The first user terminal 110 may convert the received input language voice into input language text (112). Speech-to-text conversion can be performed using any suitable speech recognition technique, acoustic models, and language models such as Hidden Markov Models, Dynamic Time Warping (DTW) based speech recognition, and other statistical and/or analysis. It can be implemented using technology. Those of ordinary skill in the relevant field can clearly understand the speech-to-text conversion, and a more detailed description will be omitted. In addition, although FIG. 1 shows that the first user terminal 110 converts the input language voice into the input language text, the speech-to-text conversion is separate from the first user terminal 110. It may be performed in a speech recognition engine (not shown).

The first user terminal 110 converting the input language voice into the input language text may transmit the input language text to the server 130 through a network. In addition, the first user terminal 110 may transmit information of chat participants including the second user terminal 190 to receive the chat message together with the input language text to the server 130. The information of chat participants may include identifiers of chat participants. For example, a user of the first user terminal 110 is a second user from a user list or a contact list stored in the first user terminal 110 through a touch interface and an application provided by the first user terminal 110. The terminal 190 can be selected. For example, the identifier of the second user terminal 190 may include a phone number, a serial number, an identification character, etc. assigned to the second user terminal 190. In addition, the information of the participants in the chat may include the language of the participants.

The server 130, which has received the input language text and the chat participants information, may determine a target language, which is a language used by users of the second user terminal 190, based on the chat participants information. For example, the first user terminal 110 may transmit an identifier of a target language to which the input language text is to be translated to the server 130 together with the input language text. The target language means a different language when the text of the input language is to be translated into the text of another language. For example, when the target language is German, the text "Hello" in the input language may be translated into the text "Hallo" in the target language. The server may receive selected information from the user and identify the input language and the target language.

Further, according to another embodiment of the present invention, the first user terminal 110 may receive an input for selecting a target language prior to receiving the input of the input language text. For example, the user may select a target language in which the input language text is to be translated through a touch interface and an application provided by the first user terminal 110.

The server 130 that has determined the target language may generate an optimal translation path for translation in connection with the translation engines 150 (131). The server 130 is a server that processes operations for generating a translation path, and may be implemented as a software module, a hardware module, or a combination thereof. The server 130 may execute an application or program for performing a translation method or a method for generating a translation path, and load an application program recorded in an internal memory. The application program recorded in the internal memory can be executed through wireless communication with the server.

The server 130 may be a gateway server that provides a translation service in connection with the translation engines 150. The server 130 may generate an optimal translation path corresponding to an input language and a target language so as to provide an optimal translation quality based on translation information for each language of the translation engines 150.

The translation engines 150 may be subjects that process an operation of translating text of a specific language into text of another language, and may include translation engines provided for each country and translation engines developed to provide translation services. have. The method of generating an optimal translation path is described in detail in FIGS. 2-6 below.

The server 130 may transmit an optimal translation path to the translation engines 150 (132). The server 130 may receive the target language text of the target language in which the input language text is translated from the translation engines 150 (132, 151). The server 130 may store the target language text (133) and transmit the target language text to the speech synthesis engines 170 (134).

The speech synthesis engines 170 may convert the target language text into the target language speech (171). A speech synthesis engine to perform text-to-speech conversion may be determined based on speech synthesis information for each language of the speech synthesis engines 170.

The target language voice generated by the speech synthesis engine may be transmitted to the server 130, and the server 130 may store the target language voice (135 ). The server 130 that has obtained the target language text and the target language voice may transmit the target language text and the target language voice to the second user terminal 190 (136). In addition, the server 130 may also transmit the target language text and the target language voice to the first user terminal 110 (137).

The second user terminal 190 may display the input language text and/or the target language text (191). For example, the second user terminal 190 may display the input language text and/or the target language text in one speech balloon. My

The first user terminal 110 may also display the input language text and/or the target language text (114). The user of the first user terminal 110 may also simultaneously view the input language text input by the user and the target language text in which the input language text is translated.

In the display according to the exemplary embodiment, it has been described that the input language text and the target language text are displayed at the same time, but the user of the first user terminal 110 or the user of the second user terminal 190 is the first language of the first language. Text to be displayed may be selected from among text and second text in a second language. For example, the user may select a text to be displayed among input language text and target language text through a user interface displayed on the terminals 110 and 190. For another example, the user may select at least one of an input language text and a target language text by touching a speech bubble displayed on the user terminals 110 and 190. For example, each time the user touches the speech balloon once, it may be displayed in the speech balloon in the order of input language text, target language text, input language text + target language text. The user terminals 110 and 190 may select the type of language of the text displayed in the speech bubble according to the user's input.

The second user terminal 190 may reproduce a target language voice (192).

According to the multilingual chat system 100, a user of the first user terminal 110 and a chat sender can transmit a voice or text message in their input language, and a user of the second user terminal 190 and a chat recipient are themselves A target language text and a target language voice translated into a target language to be used may be received.

2 is a diagram illustrating a server for translation according to an embodiment.

Referring to FIG. 2, the server 201 may generate an optimal translation path for translation in connection with the translation engines 202. The server 201 is a server that processes operations for generating a translation path, and may be implemented as a software module, a hardware module, or a combination thereof. The server 201 may execute an application or program for performing a translation method or a method for generating a translation path, and load an application program recorded in an internal memory. The application program recorded in the internal memory can be executed through wireless communication with the server.

The translation engines 202 may be subjects that process an operation of translating text of a specific language into text of another language, and include translation engines provided for each country and translation engines developed to provide a translation service. The translation engines 202 may have different translation capabilities depending on the input language and the target language. Information including the input language of the translation engines 202 and translation performance according to the target language may be referred to as translation information for each language of the translation engines 202. The translation information for each language of the translation engines 202 includes at least one of a translation accuracy for each language, a translation time corresponding to the translation accuracy for each language of the translation engines 202, and a regulation for each language of the translation engines 202. Can include. For example, translation information between language 1 and language 2 of the translation engine A may include translation accuracy, translation time required, and whether or not to regulate the translation of language 1 to language 2. Whether or not to be regulated may mean whether or not to use a translation engine in a specific country for reasons such as national policy. For example, the use of translation engine A in country 1 can be prohibited by policy. The translation information may be a score determined by a predetermined weight of each element constituting the translation information.

Although not shown, the server 201 may include a database, a processor, and a communication module. Here, the database records translation information for each language and chat participant information of the translation engines 202, the processor executes a program, and can control the server 201, and the communication module is an input language, target language, and input The text of the language and the text of the target language may be received. For example, the communication module receives the text of the input language and the target language from the user terminal 203 and transmits it to the translation engines 202, and receives the text of the target language from the translation engines 202 to the user terminal ( 203). The user terminal 203 may be connected to the server 201 through wired or wireless communication to execute an application or program for translation. The user terminal 203 may receive an input language voice and transmit it to the server 201, and may receive and output an interpreted language voice from the server 201.

The server 201 may generate an optimal translation path specified by at least one optimal translation engine based on translation information for each language of the translation engines 202 recorded in the database. The server 201 may provide an automatic interpretation portal service that generates an optimal translation path by linking between different translation engines, and translates the text of the received input language into the text of the target language through the optimal translation path. have. The server 201 may be a gateway server that provides a translation service in connection with the translation engines 202. The server 201 may generate an optimal translation path corresponding to an input language and a target language so as to provide an optimal translation quality based on translation information for each language of the translation engines 202.

The server 201, which receives the input language text and determines the target language, based on the translation information for each language of the translation engines 202, the input language of the translation engines 202 and first translation information for the target language. Can be obtained. The server 201 may obtain the input language of the translation engines 202 and second translation information for a plurality of languages based on the translation information for each language of the translation engines 202. The server 201 may obtain third translation information for a target language and a plurality of languages of the translation engines 202 based on the translation information for each language of the translation engines 202. For example, when the input language is Korean and the target language is German, the server may select languages other than German from among a plurality of languages recorded in the database, and obtain third translation information between the selected languages and German.

The server 201 may generate an optimal translation path of the input language text based on the first translation information, the second translation information, and the third translation information. Generating the optimal translation path may be determining at least one optimal translation engine corresponding to the optimal translation path.

The server 201 may generate an optimal translation path through the translation engine B and the translation engine C among the translation engines 202 in order to translate the text of the input language Korean into the text of the target language, Vietnamese. For example, the server 201 combines the translation path of "Korean -> English" through the translation engine B and the translation path of "English -> Vietnamese" through the translation engine C, so that "Korean -> English -> Vietnamese" To create an optimal translation path of ". The server 201 may determine the translation engine B and the translation engine C as optimal translation engines. The relay language refers to a language between an input language and a target language among a sequence of languages constituting a translation path. The relay language may include at least one or more languages.

3 is a flowchart illustrating a method of generating an optimal translation path according to an exemplary embodiment.

Referring to FIG. 3, a method of generating an optimal translation path according to an embodiment may be performed by the server 201 of FIG. 2. The server is a server that processes operations for translation, and may be implemented as a software module, a hardware module, or a combination thereof. The server may execute an application or program for performing the translation method, and load the application program recorded in the internal memory. The application program recorded in the internal memory can be executed through wireless communication with the server.

The server 201, based on the input language of the translation engines 202, second translation information for a plurality of languages, and third translation information for a plurality of languages and a target language, a plurality of languages At least one relay language may be selected from (310). At least one relay language may be selected based on a combination of the second translation information and the third translation information of the translation engines 202. For example, when translation information is expressed as a score, the translation information may be defined as a product of translation information between languages in a sequence of languages included in a translation path. The server may obtain recorded translation information between a plurality of languages and calculate translation information of a plurality of translation paths. The plurality of translation paths may be a sequence of "input language -> at least one relay language -> target language". According to an embodiment, the plurality of languages may include languages other than a target language among languages recorded in a database. For example, when the input language is Korean and the target language is German, the server may select languages other than German from among a plurality of languages recorded in the database, and obtain second translation information between Korean and the selected languages.

The server 201 may compare the input language, at least one relay language, and the fourth translation information of the translation path, which is a sequence of the target language, and the first translation information for the input language and the target language (320). The translation path may be defined as a sequence of languages for translating the text of the input language into the text of the target language, and the optimal translation path means an optimized translation path generated by the server. For example, when the input language is Korean and the target language is German, the translation path may be expressed as a sequence of languages from an input language such as "Korean -> English -> French -> Spanish -> German" to the target language. .

The server 201 may determine one of a translation path corresponding to the first translation information and a translation path corresponding to the fourth translation information as the optimal translation path based on the comparison result (330). For example, when the fourth translation information is higher than the first translation information, a sequence of "input language -> at least one relay language -> target language" may be determined as an optimal translation path, and the first translation information is the fourth If it is higher than the translation information, the sequence of "input language -> target language" may be determined as an optimal translation path. In addition, determining the optimal translation path by the server 201 may mean determining an optimal translation engine corresponding to the optimal translation path.

4 is a diagram illustrating translation information for each language of translation engines according to an exemplary embodiment.

The server according to an embodiment may obtain first translation information between an input language and a target language from a pre-built database. The database may be implemented as a memory included in the server, or may be implemented as an external device such as a server that can be connected to the server by wire, wireless, or network.

Translation information (A) for each language of the translation engines may be recorded in a database as shown in FIG. 4. The server may identify the input language and the target language, and obtain first translation information between the input language of the translation engines and the target language from the database. For example, if the input language is Korean and the target language is English, the server will send the translation information A ₁ (1, 2), A ₂ (1, 2) and A of each of the translation engines K ₁ , K ₂ and K ₃ from the database. ₃ (1, 2) can be obtained. A ₁ (1, 2) may be translation information of the translation engine K ₁ when Korean is the input language and the target language is English. Alternatively, if the input language is Vietnamese and the target language is English, the server sends the translation information A ₁ (N, 2), A ₂ (N, 2) and A ₃ (each of the translation engines V ₁ , V ₂ and V ₃₎ from the database. N, 2) can be obtained. A ₁ (N, 2) may be translation information of the translation engine K ₁ when Korean is the input language and the target language is English. Translation information for each language of the translation engines may be recorded in a database, and an optimal translation path and optimal translation engines may be determined based on the translation information for each language of the translation engines recorded in the database.

The translation information may be a parameter regarding translation performance between two languages. The translation information for each language of the translation engines may be numerical information including at least one of a translation accuracy for each language of the translation engines, a translation time corresponding to the translation accuracy for each language of the translation engines, and a regulation for each language of the translation engines. As described above, the manner in which translation information is defined may be applied by applying various techniques according to design intent. For example, it may be a score determined by a predetermined weight of each element constituting the translation information. The translation information may be set to different values depending on the direction of translation (eg, first language -> second language or second language -> first language), or may be defined as a value independent of the direction of translation. In addition, the unit can be variously modified and defined according to the design intent, and can be expressed in %, for example.

According to an embodiment, the translation accuracy between the first language and the second language is determined by the amount of translation data retained between the first language and the second language, the word order match rate between the first language and the second language, and the character of the first language. It can be defined based on at least one factor among the matching of the capacity and the capacity of the characters in the second language, and the feedback corresponding to the translation between the first language and the second language, and the accuracy of the translation as these factors are updated. Can be updated. Here, the capacity of a character means a capacity required to express a character of a specific language. For example, since Korean Hangul can be expressed in 2 bits and English alphabet can be expressed in 1 bit, the capacity of Korean and alphabets do not match. Translation accuracy can be defined by weighting factors affecting the translation accuracy. For example, a weight of a holding amount of translation data and a weight of a word order match rate may be set differently to define translation accuracy. As described above, the manner in which translation accuracy is defined may be applied by applying various techniques according to design intention.

According to an embodiment, the translation accuracy may be an index related to the performance of a translation engine that provides translation between specific languages. Here, the translation engine may be an independent server or device separate from the server according to an embodiment. The server may transmit the text of the input language to the translation engine and receive the text of the target language as a result of the translation from the translation engine. The translation accuracy recorded in the database may be the translation accuracy of a plurality of translation engines. In this case, the server may process an operation of selecting translation engines for translating the text of the input language into the text of the target language based on the translation accuracy recorded in the database. Translation accuracy can be obtained from the execution entity of the translation engine and recorded in a database.

According to an embodiment, the database may record the required translation times corresponding to the translation accuracy. The server may select translation engines for translating the text of the input language into the text of the target language, taking into account the translation accuracy and time required for translation.

According to an embodiment, the database may record whether or not translation engines are regulated. The server may select translation engines in consideration of whether or not the translation engines are regulated.

The server may select a translation path having the highest translation information among translation paths corresponding to a plurality of languages of the translation engines, and determine a language belonging to the selected translation path as a relay language. For example, when the input language is Korean and the target language is German, translation paths corresponding to a plurality of languages of translation engines may include the following examples.

Translation Path 1: "Korean -> Japanese -> English -> German"

Translation Path 2: "Korean -> Japanese -> German"

Translation Path 3: "Korean -> English -> German"

Translation Path 4: "Korean -> English -> Spanish -> German"

In the above examples, if the translation information of the translation engines of the translation path 1 is the highest, the server selects the translation path 1 and determines "Japanese" and "English", which are languages belonging to the selected translation path 1, as the relay language. I can. Here, the translation path information of the translation engines is defined based on translation information between the input language and at least one relay language, translation information between at least one relay language, and translation information between the at least one relay language and the target language. Can be.

For example, information on a translation path for each language of translation engines may be defined as a product of translation information between languages in a sequence of languages included in the translation path. The server may obtain translation information between a plurality of languages recorded as shown in FIG. 5 and may calculate translation information of a plurality of translation paths. The plurality of translation paths is a sequence of "input language -> at least one relay language -> target language". The server selects a translation path having the highest translation information among the calculated translation information. When the translation path having the highest translation information is a sequence of "input language -> at least one relay language -> target language", the server may determine a language belonging to this sequence as the relay language. When the translation information of the sequence of "input language -> target language" is the highest, the relay language may not be selected.

5 is a diagram illustrating a process of generating an optimal translation path according to an embodiment.

The server may determine one of a translation path that is a sequence of "input language -> at least one relay language -> target language" and a translation path of "input language -> target language" as the optimal translation path based on the translation information. have. The translation information of the translation path, which is a sequence of "input language -> target language", may be first translation information, and the translation information of the translation path which is the sequence of "input language -> at least one relay language -> target language" 4 May be translation information.

As described above, the fourth translation information may be defined as a product of translation accuracy between languages belonging to a sequence of a translation path. The server may compare the first translation information and the fourth translation information, and determine any one of a translation path corresponding to the first translation information and a translation path corresponding to the fourth translation information as an optimal translation path based on the comparison result. .

5, the server optimally translates a translation path 502 which is a sequence of "Korean -> Japanese -> English -> German" among translation paths for translation between a Korean input language and a German target language. You can decide by path. The server may select a translation path 502 having the highest translation information among the translation paths, and determine "Japanese" and "English" belonging to the selected translation path 502 as a relay language.

Translation information of the selected translation path may be referred to as fourth translation information, and translation information of the translation path 501 of "Korean -> German" may be referred to as first translation information. The server compares the fourth translation information with the first translation information, and as a result of the comparison, since the fourth translation information is larger than the first translation information, the translation path 502 may be determined as an optimal translation path. The translation information for "Korean -> German" is 50%, the translation information for "Korean -> Japanese" is 95%, the translation information for "Japanese -> English" is 90%, and the translation for "English -> German" Since the information is 88%, the first translation information is 50%, and the fourth translation information is 75.24% (which can be calculated as 0.95 * 0.9 * 0.88).

Here, the translation information is calculated as the product of the translation information between each language, but is not limited to such a calculation technique, and techniques such as a method of differentially applying weights or imposing a penalty as a path lengthens can be applied in various ways. If the translation performance when directly translating from the input language to the target language is better than the translation information of the translation path, which is a sequence including the relay language, or the time required for translation, the server determines the translation path that is the sequence of the input language and the target language. It can be determined as the optimal translation path.

The server may generate an optimal translation path 502 and use the generated optimal translation path 502 to translate Korean text into German text. Here, the translation operation of converting text into text of another language may be directly processed by the server or may be processed by an external server or device for each language belonging to the optimal translation path 502. Alternatively, the server may access the integrated cloud server, and translation operations according to the sequence of the optimal translation path 502 may be processed on the cloud server.

For example, the server may generate an optimal translation path 502 suitable for an input language in Korean and a target language in German, based on translation information recorded in the database and translation time required. The server searches for translation engines recorded from the database based on the generated optimal translation path 502, and uses the search result to translate "Japanese -> English", a translation engine suitable for translation of "Korean -> Japanese" It is possible to determine which translation engine is suitable for and for the translation of "English -> German". In this case, the translation between languages belonging to the optimal translation path 502 is processed through a translation engine separate from the server, and the server generates the optimal translation path 502 and relays the translated texts. have.

According to an embodiment, the server determines whether to perform translation processing through a program recorded in an internal memory or by employing an external translation engine, based on translation information recorded in the database and translation time required. You can judge whether or not. If the performance of the translation processing through the program recorded in the internal memory is superior to the translation information or the translation time required by the external server or the translation engine, the server can process the translation operation in a local manner.

According to an embodiment, when translation information (first translation information) between the input language and the target language exceeds threshold information, the server may determine the sequence of the input language and the target language as an optimal translation path. For example, if the translation information between the input language Korean and the target language Japanese is 95%, and the critical information is 90%, the server determines the sequence of "Korean -> Japanese" as the optimal translation path, and an additional translation path Or, it may not perform a search for a relay language.

According to an embodiment, the server may select languages in which translation information with an input language and translation information with a target language are in a critical information range from among a plurality of languages, and select at least one relay language among the selected languages. have. For example, the server may select languages in which translation information is in the range of 60% to 100% (or exceeds 60%) among translation information between Korean as an input language and a plurality of languages. The server may generate translation paths including selected languages, and the generated translation paths may compare translation information with each other.

According to an embodiment, when the translation information between the languages recorded in the database and the input language is less than the threshold information, or when the translation information between the languages recorded in the database and the target language is less than the threshold information, the "input language- The sequence of "target language" can be determined as an optimal translation path. For example, if the input language is Vietnamese and the target language is Indonesian, translation information between Vietnamese and multiple languages is less than 60%, and translation information between multiple languages and Indonesian is less than 60%, the server is "Vietnamese -> The translation path can be determined by the sequence of "Indonesian".

According to an embodiment, the server may derive an optimal translation path using the following algorithm.

<Algorithm>

Here, A(i, j) is translation information between language i and language j, N is the number of languages, and i -> j is a command for generating an optimal translation path for translation from language i to language j. , 1node is a command to calculate the translation information R1 of the translation path, which is a sequence of "language i -> language j", and 2node is the translation information R2 of the translation path, which is a sequence of "language i -> language k -> language j" It is an instruction to calculate, and 3node is an instruction to calculate translation information R3 of a translation path which is a sequence of "language i -> language k -> language l -> language j". The server may determine a translation path corresponding to the largest translation information among the translation information of R1 to RM as an optimal translation path. The above algorithm is only an example, and the server can derive an optimal translation path by employing algorithms such as dynamic programming and tree based optimal path search.

According to an embodiment, the server may generate an optimal translation path based on translation information between languages belonging to the translation path as well as translation times corresponding to the corresponding translation information. Here, the weights applied to the translation information and the required translation times may be variously modified according to the design intention.

6 is an exemplary diagram of a configuration of a server according to an embodiment.

Referring to FIG. 6, the server 601 includes a processor 602 and a memory 603. The processor 602 may include at least one of the devices described above with reference to FIGS. 1 to 5, or may perform at least one method described above with reference to FIGS. 1 to 5. The memory 603 may store a program in which the translation method is implemented. The memory 603 may be a volatile memory or a nonvolatile memory.

The processor 602 can execute a program and control the server 601. The code of a program executed by the processor 602 may be stored in the memory 603. The server 601 is connected to an external device (eg, a personal computer or a network) through an input/output device (not shown), and may exchange data.

In order to provide a translation between an input language and a target language, the server according to an embodiment may generate an optimal translation path or select a relay language in consideration of translation information recorded in a database and time required for translation. Through this, an exemplary embodiment can increase the quality of a translation service with a large gap between languages, and generate a translation path optimized for a language in which the quality of the translation service is poor due to a poor IT environment. In addition, in an embodiment, when translation information between specific languages is higher than that of other languages, a translation path through the corresponding languages is generated, and through this, translation processing of better quality than a direct translation path between the input language and the target language Can provide. Since raising the quality of translation between all languages is limited in terms of the amount of translation data or processing speed, one embodiment can break the language barrier between languages by creating an optimal translation path suitable for the input language and the target language. . In addition, one embodiment can apply a real-time translation service to various fields. For example, one embodiment can be applied in various ways, such as a technique for replacing an interpreter or inserting a subtitle of a foreign broadcast in real time.

7 is a flowchart illustrating a method of operating an input language terminal according to an embodiment.

Referring to FIG. 7, a method of operating an input language terminal according to an embodiment may be performed by an input language terminal. The input language terminal may be a chat sender terminal.

The input language terminal may receive an input language voice from the chat sender (710). The input language terminal may receive a voice of an input language using a voice recognition device.

The input language terminal may acquire conversion information for each language of the conversion engines (720).

The input language terminal may convert the input language voice into the input language text based on conversion information for each language of the conversion engines (730 ). The input language terminal may display the input language text (740 ).

According to another embodiment, the input language voice-to-input language text conversion may be performed in a server other than an input language terminal. When the input language voice to input language text conversion is performed in the server, the input language terminal may transmit only the input language voice to the server, and receive and display the input language text from the server.

The input language terminal may transmit information of multilingual chat participants, input language voice, and input language text to the server.

The input language terminal may convert the input language voice into the input language text based on the artificial neural network.

The input language terminal may obtain location information of the sender of the chat. The location information of the input language terminal is information indicating a location of a specific place based on a location where the input language terminal currently exists and a specific area, and may include latitude information, longitude information, address information, and the like. Location information may be typically obtained using a GPS (Global Positioning System) module, but is not limited thereto, and location information may be obtained using any device capable of determining the current location information of the input language terminal. For example, the location information can be obtained through a Bluetooth communication module, an infrared communication module, a Zigbee communication module, a WiFi communication module, a Radio Frequency IDentification (RFID) communication module, and the like. .

Even in the same language, there may be differences in dialects and dialects by country/region. For example, in the case of English, there may be slight differences depending on the region (US, UK, Australia, Philippines, etc.). Location information can be used to provide better speech recognition, translation and interpretation.

When converting the input language voice into the input language text, the input language terminal may further consider location information to provide optimal voice recognition for each country/region. In addition, the input language terminal may transmit location information to the server, and the server may determine a translation path and a speech synthesis path in consideration of the location information.

8 is a flowchart illustrating a method of operating a target language terminal according to an embodiment.

Referring to FIG. 8, a method of operating a target language terminal according to an embodiment may be performed by the target language terminal. The target language terminal may be a chat recipient terminal. The target language terminal and the input language terminal are classified according to a chat recipient and a sender, and may be an input language terminal when one terminal sends a chat, and a target language terminal when a chat is received.

The target language terminal may receive the target language text from the server (810).

The target language terminal may display the target language text (820). In addition, the target language terminal may display the source language text together with the target language text.

The target language terminal may obtain synthesis information for each language of the synthesis engines (830 ). The synthesis engines 150 may be subjects that process an operation of synthesizing text into speech, and may include synthesis engines provided for each country and synthesis engines developed to provide a synthesis service. The synthesis information for each language of the synthesis engines may be information including at least one of a synthesis accuracy for each language, a synthesis time corresponding to the synthesis accuracy, and a regulation for each language of the synthesis engines.

The target language terminal may synthesize the target language text into the target language speech based on the synthesis information for each language of the synthesis engines (840 ). The target language terminal may synthesize a target language text into a target language speech based on a synthesis engine having the highest synthesis information among the synthesis engines.

The target language terminal may reproduce the target language voice (850).

If the input language and the target language are the same, the target language terminal may not receive the target language text, but may receive only the target language voice.

The embodiments described above may be implemented with hardware components, software components, and/or combinations of hardware components and software components. For example, the devices, methods, and components described in the embodiments include, for example, processors, controllers, arithmetic logic units (ALUs), digital signal processors (micro signal processors), microcomputers, and field programmable gates (FPGAs). It can be implemented using one or more general purpose computers or special purpose computers, such as arrays, programmable logic units (PLUs), microprocessors, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications executed on the operating system. In addition, the processing device may access, store, manipulate, process, and generate data in response to the execution of the software. For convenience of understanding, a processing device may be described as one being used, but a person having ordinary skill in the art, the processing device may include a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that may include. For example, the processing device may include a plurality of processors or a processor and a controller. In addition, other processing configurations, such as parallel processors, are possible.

The software may include a computer program, code, instruction, or a combination of one or more of these, and configure the processing device to operate as desired, or process independently or collectively You can command the device. Software and/or data may be interpreted by a processing device or to provide instructions or data to a processing device, of any type of machine, component, physical device, virtual equipment, computer storage medium or device. , Or may be permanently or temporarily embodied in the transmitted signal wave. The software may be distributed on networked computer systems, and stored or executed in a distributed manner. Software and data may be stored in one or more computer-readable recording media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, or the like alone or in combination. The program instructions recorded on the medium may be specially designed and constructed for the embodiments or may be known and usable by those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs, DVDs, and magnetic media such as floptical disks. -Hardware devices specifically configured to store and execute program instructions such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include high-level language code that can be executed by a computer using an interpreter, etc., as well as machine language codes produced by a compiler.

As described above, although the embodiments have been described by the limited drawings, those skilled in the art can apply various technical modifications and variations based on the above. For example, the described techniques are performed in a different order than the described method, and/or the components of the described system, structure, device, circuit, etc. are combined or combined in a different form from the described method, or other components Alternatively, even if replaced or substituted by equivalents, appropriate results can be achieved.

Therefore, other implementations, other embodiments, and claims and equivalents fall within the scope of the claims to be described later.

Claims (21)

Receiving information on multilingual chat participants;
Receiving input language text from the chat sender;
Determining a target language for each of the chat recipients based on the information of the multilingual chat participants;
Obtaining translation information for each language of the translation engines;
Obtaining first translation information for the input language and the target language of the translation engines based on the translation information for each language of the translation engines;
Obtaining second translation information for the input language and a plurality of languages of the translation engines based on the translation information for each language of the translation engines;
Acquiring the plurality of languages of the translation engines and third translation information for the target language based on the translation information for each language of the translation engines; And
Generating an optimal translation path of the input language text based on the first translation information, the second translation information, and the third translation information
Containing
How the server operates in multilingual chat.
The method of claim 1,
Generating the optimal translation path,
Determining at least one optimal translation engine corresponding to the optimal translation path
Containing,
How the server operates in multilingual chat.
The method of claim 1,
Translation information for each language of the translation engines
Including at least one of translation information for each language of the translation engines, a translation time corresponding to the translation information for each language of the translation engines, and regulations for each language of the translation engines,
How the server operates in multilingual chat.
The method of claim 1,
Generating the optimal translation path
Selecting at least one relay language from the plurality of languages based on the second translation information and the third translation information;
Comparing the first translation information with fourth translation information of a translation path that is a sequence of the input language, the at least one relay language, and the target language; And
Based on the comparison result, determining one of a translation path corresponding to the first translation information and a translation path corresponding to the fourth translation information as the optimal translation path
Containing,
How the server operates in multilingual chat.
The method of claim 1,
Receiving location information of the chat sender
Further comprising,
Translation information for each language of the translation engines
Including the location information of the chat sender,
How the server operates in multilingual chat.
The method of claim 1,
The information of the participants includes the language of the participants,
How the server operates in multilingual chat.
According to claim 2,
Translation information between the first language and the second language
The amount of translation data held between the first language and the second language, a word order matching rate between the first language and the second language, whether the capacity of the first language and the second language match And updated based on at least one of feedback corresponding to the translation between the first language and the second language,
The first language and the second language include the input language, the target language, and the at least one relay language,
How the server operates in multilingual chat.
The method of claim 1,
Receiving a target language text translated into a target language of each of the chat recipients based on the optimal translation path; And
Transmitting each of the target language texts to the chat recipients
Further comprising,
How the server operates in multilingual chat.
The method of claim 1,
The input language text is
Containing text converted to the voice of the input language,
How the server operates in multilingual chat.
Receiving an input language voice from a chat sender;
Obtaining conversion information for each language of the conversion engines;
Converting the input language speech into input language text based on the conversion information for each language of the conversion engines;
Displaying the input language text; And
Transmitting information of multilingual chat participants, the input language voice, and the input language text to a server
Including,
The server
Determine a target language of each of the chat recipients based on the information of the multilingual chat participants,
Generating an optimal translation path of the input language text based on translation information for each language of the translation engines
Operation method of input terminal in multilingual chat.
The method of claim 10,
Obtaining location information of the chat sender;
Transmitting the location information of the chat sender to a server; And
Converting the input language voice into the input language text based on the location information
Further comprising,
Operation method of input terminal in multilingual chat.
The method of claim 10,
Converting the input language speech into the input language text
Converting the input language speech to the input language text based on an artificial neural network
Containing,
Operation method of input terminal in multilingual chat.
Receiving a target language text from a server;
Displaying the target language text;
Obtaining synthesis information for each language of the synthesis engines;
Synthesizing the target language text into a target language speech based on the synthesis information for each language of the synthesis engines; And
Reproducing the target language voice
Including,
The server
Determine the target language of each of the chat recipients based on the information of the multilingual chat participants,
Generates the optimal translation path of the input language text based on the translation information of each language of the translation engines.
Operation method of target terminal in multilingual chat.
The method of claim 13,
When the input language and the target language are the same, receiving only the target language voice without receiving the target language text
Further comprising,
Operation method of target terminal in multilingual chat.
A computer program stored in a computer-readable recording medium to execute the method of any one of claims 1 to 14 in combination with hardware.
In the server for multilingual chat,
A database for recording translation information of each language of translation engines and information of multilingual chat participants; And
Processor that creates an optimal translation path between the input language and the target language.
Including,
The processor
Obtaining first translation information for the input language and the target language of the translation engines based on the translation information for each language of the translation engines,
Acquiring second translation information for the input language and a plurality of languages of the translation engines based on the translation information for each language of the translation engines,
Obtaining third translation information for the target language and the plurality of languages of the translation engines based on the translation information for each language of the translation engines,
Determining at least one optimal translation engine based on the first translation information, the second translation information and the third translation information,
Generating an optimal translation path of the input language text based on the at least one optimal translation engine,
server.
The method of claim 16,
The processor
Determining at least one optimal translation engine corresponding to the optimal translation path,
server.
The method of claim 16,
Translation information for each language of the translation engines
Including at least one of translation information for each language of the translation engines, a translation time corresponding to the translation information for each language of the translation engines, and regulations for each language of the translation engines,
server.
The method of claim 16,
The processor
Based on the second translation information and the third translation information, a translation that is a sequence of the input language, the at least one relay language, and the target language by selecting at least one relay language from the plurality of languages The fourth translation information of the path and the first translation information are compared, and based on the comparison result, one of a translation path corresponding to the first translation information and a translation path corresponding to the fourth translation information is optimized. Determined by the translation path,
server.
The method of claim 16,
The processor
Receive the location information of the sender of the chat,
Translation information for each language of the translation engines
Including the location information of the chat sender,
server.
The method of claim 16,
The information of the participants includes the language of the participants,
server.

Images