KR102170684B1 - System for interpreting and translating using smart clothes - Google Patents

Abstract

The present invention includes an input unit for collecting voices spoken between chatters and converting them into voice signals; And it discloses a smart clothing interpretation and translation system including a smart clothing including an output unit for outputting at least one form of a text and a synthesized voice translated content for the voice content recognized using the voice signal. According to the present invention, it is possible to output the result of interpretation and translation through audio and video using smart clothing.

Description

Interpretation and translation system using smart clothing {SYSTEM FOR INTERPRETING AND TRANSLATING USING SMART CLOTHES}

The present invention relates to an interpretation and translation system using smart clothing, and more particularly, to an interpretation and translation system using smart clothing that recognizes voices between talkers or between speakers and audiences and outputs translated and interpreted content through smart clothing. About.

A wearable device refers to all electronic devices that are attached to a body to perform computing actions, and includes applications that can perform some computing functions.

In addition, a wearable device refers to a next-generation electronic device that is developed small and light so that it can be worn on a body or clothing so that the user can freely use it while moving or performing activities, and thus can communicate with the user at the nearest place of the body.

Wearable devices are largely classified into accessory type, clothing integrated type, and body attachment type. Among them, wearable devices with integrated clothing are called smart clothing.

The interpretation and translation system using smart clothing according to an embodiment of the present invention combines smart clothing that has been limitedly used for biometrics or context-based color/pattern change with an interpretation and translation algorithm, and input and input through smart clothing between talkers or between speakers and audiences. The present invention relates to an interpretation and translation system for performing audio and video interpretation and translation using an output function, and is a technology that is distinguished from the above-described conventional technology, and is to solve the above problem.

The present invention has been created to solve the above problems, and an object of the present invention is to provide an interpretation and translation system using smart clothing capable of outputting interpretation and translation results through audio and video using smart clothing.

In addition, it is an object of the present invention to provide an interpretation and translation system using smart clothing capable of outputting interpretation and translation results through a remote interpretation and translation server using a communication function of a user terminal connected by wire or wirelessly.

In addition, an object of the present invention is to provide an interpretation and translation system using smart clothing in which the sizes of voices and texts that are output are automatically adjusted according to a distance between a speaker or a speaker and an audience.

In addition, an object of the present invention is to provide an interpretation and translation system using smart clothing that can recognize external noise and actively remove it.

An interpretation and translation system using smart clothing according to an embodiment of the present invention includes: an input unit for collecting voice for conversion from spoken voice to voice signal; And a smart clothing including an output unit that outputs the translated content of the voice content recognized using the voice signal in at least one form of text and synthesized voice.

Here, it characterized in that it further comprises an interpretation and translation unit for performing translation and interpretation through a speech recognition process using the voice signal.

Here, the interpretation and translation unit is characterized in that it is implemented in the form of a user terminal that is connected to the smart clothing by wire or wirelessly to translate and interpret the spoken voice.

Here, the smart clothing includes: a distance sensor for recognizing a distance from a user to a talker or an audience; And a controller for controlling a volume of a voice and a volume of a text output through the output unit using the sensing information of the distance sensor.

Here, the smart clothing further comprises an audio processing unit configured to include a plurality of microphones of the input unit over the front and rear surfaces, and to recognize and filter a voice signal using a volume difference of the voice according to the location of the microphone. To do.

Here, the smart clothing analyzes the waveform of the noise signal included in the voice signal, generates an interference wave that cancels the analyzed waveform, and receives the synthesized wave of the noise signal and the interference wave as feedback so that the synthesized wave is canceled. It characterized in that it further comprises a noise processing unit for correcting the interference wave.

Here, the smart clothing includes: a power supply unit for supplying power to the input unit and the output unit; And a charging unit for charging the user terminal using the power supply unit.

Here, the output unit includes a display device, and the display device is positioned at least one of the front of the chest and the abdomen and the rear of the back of the smart clothing so that others other than the user can see it, so that the user can see, It characterized in that it is located on the arm of the smart clothing.

According to an embodiment of the present invention, a method for interpreting and translating using smart clothing includes the steps of: collecting spoken speech through an input unit of the smart clothing and converting it into a speech signal; Performing translation and interpretation through a voice recognition process using the voice signal; And outputting the translation and interpretation results in at least one form of text and synthesized voice using smart clothing.

Here, the step of performing the translation and interpretation may be performed by a user terminal that is connected to and communicates with the smart clothing by wire or wirelessly.

Here, the interpretation and translation method using the smart clothing includes the steps of recognizing a voice signal using a difference in volume of a voice according to a location of a microphone and filtering the voice signal; And analyzing the waveform of the noise signal included in the voice signal, generating an interference wave that cancels the analyzed waveform, and correcting the interference wave so that the synthesized wave is canceled by receiving feedback of the synthesized wave of the noise signal and the interference wave. It characterized in that it further comprises at least one of the steps.

According to the present invention, it is possible to output the result of interpretation and translation through audio and video using smart clothing.

In addition, it is possible to output an interpretation and translation result through a service of a remote interpretation/translation server using a communication function of a user terminal connected by wire or wirelessly.

In addition, an interpretation and translation result may be output so that the size of the voice and text output is automatically adjusted according to the distance between the talker or between the speaker and the audience using smart clothing.

In addition, external noise can be recognized and actively removed through smart clothing.

1 is an exemplary diagram of a network environment of an interpretation and translation system using smart clothing according to an embodiment of the present invention.
2 is a block diagram of an interpretation and translation server according to an embodiment of the present invention.
3 is a block diagram of smart clothing according to an embodiment of the present invention.
4 is an exemplary diagram of smart clothing according to an embodiment of the present invention.
5 is an exemplary diagram of an interpretation and translation method using smart clothing according to an embodiment of the present invention.
6 is an exemplary diagram of a preprocessing step of a voice signal according to an embodiment of the present invention.

Hereinafter, a preferred embodiment of an interpretation and translation method and an interpretation and translation system using smart clothing will be described in detail with reference to the accompanying drawings. The same reference numerals in each drawing indicate the same member. In addition, specific structural or functional descriptions of the embodiments of the present invention are exemplified only for the purpose of describing the embodiments according to the present invention, and unless otherwise defined, all terms used herein including technical or scientific terms They have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs. 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 this specification. It is desirable not to.

Hereinafter, a network environment of an interpretation and translation system using smart clothing according to an embodiment of the present invention will be described.

1 is an exemplary diagram of a network environment of an interpretation and translation system using smart clothing according to an embodiment of the present invention.

Referring to FIG. 1, an interpretation and translation system 1 using smart clothing according to an embodiment of the present invention is shown. The interpretation and translation system 1 using smart clothing basically includes the smart clothing 100, and may additionally include a user terminal 200 and an interpretation and translation server 300.

The smart clothing 100 receives and recognizes foreign language voices between talkers or between audiences and audiences, performs translation and interpretation using the recognized contents, and outputs the results of speech synthesis or text. Functions. Here, the interpretation and translation system 1 using smart clothing may include only the smart clothing 100 or a user terminal 200 and an interpretation and translation server 300 according to which device performs voice recognition, translation, interpretation, and speech synthesis functions. Can include.

The user terminal 200 may receive a voice signal input through the smart clothing 100 and use this to perform at least one of voice recognition, translation, interpretation, and speech synthesis functions. For example, the user terminal 200 directly performs voice recognition, translation, interpretation and speech synthesis functions, or performs voice recognition, translation, and interpretation functions that require a large database through the interpretation and translation server 300, and a speech synthesis function. Can be performed alone.

The user terminal 200 includes a communication unit 210, a display unit 220, a speech synthesis unit 230, a storage unit 240, an input unit 250, an output unit 260, a playback unit 270, and a power supply unit 280. ) And a control unit 290.

Various embodiments of the user terminal 200 include a cellular phone, a smart phone having a wireless communication function, a personal portable terminal (PDA) having a wireless communication function, a wireless modem, a portable computer having a wireless communication function, and a digital device having a wireless communication function. A portable device that incorporates combinations of such functions, as well as photographing devices such as cameras, gaming devices with wireless communication functions, music storage and playback home appliances with wireless communication functions, and Internet home appliances capable of wireless Internet access and browsing. It may include a unit or terminals, but is not limited thereto.

The interpretation and translation server 300 may selectively perform voice recognition, translation, interpretation, and speech synthesis functions through sharing with the user terminal 200.

2 is a block diagram of an interpretation and translation server according to an embodiment of the present invention.

Referring to FIG. 2, the interpretation and translation server 300 includes a speech recognition module 310, a speech DB 315, a translation module 320, a translation DB 325, and a speech synthesis module 330.

The voice recognition module 310 records a talker's voice and performs voice recognition using the recorded data. The voice recognition module 310 automatically recognizes the voice signal from the talker's mouth and converts it into a character string. Another name for the speech recognition unit is automatic speech recognition (ASR), voice recognition or speech-to-text (STT).

The speech recognition module 310 may be based on a probability statistics method. That is, the speech recognition module 310 uses a model based on probability statistics as an acoustic model (AM) language model (LM) used in the speech recognition process. And the core algorithm HMM (hidden markov model) can also be based on probability statistics. The above models are examples and are not intended to limit the present invention.

A Gaussian Mixture Model (GMM) may be used as an acoustic model, and N-gram may be used as a language model. Furthermore, it is also preferable to use a Deep Neural Network (DNN), one of deep learning architectures, instead of GMM. In addition, in order to improve the performance of speech recognition, a high-quality speech model and a language model are set, and the set models can be learned by a deep learning algorithm. It is preferable that the learning DB required for learning includes spoken and conversational speech and language DB.

When the speech of the speaker uttered in the source language recognized by the speech recognition module 310 is output as text, the translation module 320 translates the output text into characters of a target language. The interpretation server 300 according to an embodiment of the present invention is characterized in that it includes a translation module 320 as well as a voice recognition module 310.

The translation method performed by the translation module 320 includes at least one of a rule-based method, a corpus-based method, and an artificial neural network translation (NMT). Rule-based methods are divided into direct translation methods, indirect conversion methods, and intermediate language methods according to the depth of analysis. There are two types of corpus-based methods, an example-based method and a statistical method.

Stochastic Machine Translation (SMT) technology is a technology that translates sentences by learning model parameters from a bilingual corpus through statistical analysis. When developing a grammar or semantic representation, a model necessary for translation is created from the corpus of the language pair to be translated rather than manually. So, if only the corpus can be secured, the language can be expanded relatively easily.

The disadvantage of statistics-based automatic translation technology is that it requires a large-scale bilingual corpus, and there is no common semantic representation that connects multiple languages.

Neural Machine Translation (NMT) is a technology to compensate for these shortcomings.

SMT is a method of dividing a sentence into words or phrases of several words and then translating it based on a statistical model. The key is to model statistical translation rules based on vast training data.

In contrast, in NMT, artificial intelligence (AI) translates the whole sentence. The reason why sentence-by-sentence translation is possible is because the artificial neural network converts sentence information into a vector (coordinate value) that means a specific point in the virtual space.

For example, the word'person' is called'[a, b, c,… , x, z]'. Since the vector contains all information such as words, phrases, and word order, it is possible to translate sentences by understanding the context. Artificial neural networks place sentences with similar meanings in close proximity to each other.

High-dimensional vectors are used in NMT technology. The artificial neural network is trained by using the learning data consisting of the sentences of the starting word and the sentence of the destination word, and the learned artificial neural network recognizes the sentence information as a vector.

The speech synthesis module 330 outputs a synthesized speech corresponding to the speaker's speech according to the translation by the translation module 310. When the smart clothing 100 performs a speech synthesis function, TTS data is received from the interpretation and translation server 300 through the user terminal 200. The smart clothing 100 synthesizes speech using a speech synthesis module based on TTS data. The smart clothing 100 reproduces the TTS data transmitted from the user terminal 200, and the result is output to a speaker.

When the interpretation and translation server 300 includes the speech synthesis module 330, the interpretation and translation server 300 itself generates TTS data.

Speech synthesis is called TTS (Text-to-speech) or Voice Synthesis. As a method of speech synthesis, a sound combination method can be used. In the sound piece combination method, the sound piece according to the sentence analysis and analysis result is extracted from the sound piece DB, and it is connected. A composite sound of several candidates is generated, and the most suitable one is selected in consideration of prosody and smoothness. Furthermore, by determining the tone of the talker using the sound spectrum of the talker's voice, and post-processing the synthesized tone to match the tone, a synthesized sound close to the tone of the primary talker can be output. In addition, the speaker's emotion is recognized, and the perceived emotion may be carried on the synthesized sound.

The speech recognition DB 315 and the translation DB 325 correspond to databases determined by reducing the recognition range in the order of lower frequency of appearance according to the type of language, and the voice recognition module 310 and the translation module 320 are miniaturized. It includes an engine that uses the voice recognition DB or translation DB.

The speech recognition DB 315 is characterized in that it is a DB constructed by learning speech due to various utterances using a deep learning algorithm and reducing or expanding a recognition range according to the frequency of speech content. In other words, to recognize high-frequency speech content, the amount of DB is relatively increased, and in order to recognize low-frequency speech content, the amount of DB is significantly reduced.

In order to obtain a high-quality voice recognition rate, the larger the amount of voice recognition DB 315 is, the more advantageous there is, but there are problems of delay and overload. Do.

Also, for the translation DB 325, similarly to the above method, various translation examples are learned using deep learning algorithms, colloquial expressions are expanded according to the frequency of translation examples, and written language expressions are reduced to build a DB. have.

Accordingly, the speech recognition module 310 and the translation module 320 according to the present invention can use a low-capacity voice recognition DB 315 or a translation DB 325 compared to a DB constructed without considering the frequency.

The network 400 is a wired and wireless network, for example, a local area network (LAN), a wide area network (WAN), an Internet, an intranet and an extranet, and a mobile network, for example. It may be any suitable communication network including cellular, 3G, LTE, WiFi networks, ad hoc networks, and combinations thereof.

Network 400 may include a connection of network elements such as hubs, bridges, routers, switches and gateways. Network 400 may include one or more connected networks, such as a multi-network environment, including a public network such as the Internet and a private network such as a secure corporate private network. Access to network 400 may be provided through one or more wired or wireless access networks.

3 is a block diagram of smart clothing according to an embodiment of the present invention.

Referring to FIG. 3, a smart clothing 100 according to an embodiment of the present invention includes an input unit 110, an output unit 120, a PCB 130, a distance sensor 140, and a charging unit 150. .

The input unit 110 performs a function of receiving a conversation content between a speaker and a voice between a speaker and an audience, and a function of receiving letters and numbers from a user. The input unit 110 includes a microphone 111 and a keyboard 112.

The output unit 120 serves to output the translated content as text and output the interpreted content as voice. The output unit 120 includes a display device 121 and a speaker 122.

The input unit 110 and the output unit 120 are located outside the smart clothing 100.

4 is an exemplary diagram of smart clothing according to an embodiment of the present invention.

Referring to FIG. 4, a microphone 111 and a speaker 122 for inputting or outputting sound may be located on a chest of smart clothing. However, since each position shown in each drawing is only one of several exemplary embodiments, it is not limited thereto.

The keyboard 112 is preferably positioned on the left arm in the case of a right-handed person taking into account the position of the user's hand. The keyboard may be implemented in a touch method using a touch sensor.

The display device 121 is positioned on at least one of the front of the chest and the abdomen and the rear of the back of the smart clothing so that others other than the user can see it. In addition, the display device 121 is positioned on the arm of the smart clothing so that the user can see it.

The PCB 130 is a mechanical component and may include various electronic components and the like therein. The PCB 130 may be implemented in the form of a film using a flexible flat cable (FFC). The PCB 130 may be located on the smart clothing 100 without any particular restrictions, and may be located on the arm or in particular.

The distance sensor 140 serves to measure the distance to the talker or the audience through infrared light or the like. The distance sensor 140 is preferably located in front of the smart clothing 100 due to its characteristics.

The charging unit 150 serves to charge the user terminal 200 by wire or wirelessly using the power supply of the smart clothing 100. The user terminal 200 may be charged by wire through a USB terminal, or may be charged wirelessly by receiving power from a transmitter that transmits wireless power while contained in a pocket of the smart clothing 100.

And the PCB 130 corresponding to the mechanical configuration includes a control unit 1310, a power supply unit 132, an interpretation and translation unit 1300, an audio processing unit 134 and a noise processing unit 135.

The control unit 131 serves to control each component of the smart clothing 100. For example, the volume of the voice output through the output unit 120 and the size of the text are controlled using the sensing information of the distance sensor 140.

The power supply unit 132 serves to supply power required for input, output, and communication functions of the input unit 110 and the output unit 120 and other smart clothing 100. As constituent elements of the power supply unit 132, a secondary battery such as Litum ion, a circuit for charging and discharging, and a protection circuit are included.

The interpretation and translation unit 133 records the input voice signal and performs voice recognition, translation, interpretation, and speech synthesis functions using the recorded voice data. The entire function of interpretation and translation by the interpretation and translation unit 133 is similar to the function performed by the interpretation and translation server 300.

The audio processing unit 134 serves to pre-process the audio signal when the interpreter and translation unit 133 performs translation and interpretation using the audio signal. A plurality of microphones 111 may be included in the smart clothing 100. Voice input according to the location of the microphone 111 shows a difference in volume. For example, when having a conversation with a talker in front of the user, a louder voice is input through the microphone 111 located in front of the user rather than in the rear. In addition, when there is noise included in the voice, the size of the noise may be input at a constant size regardless of the front or back of the smart clothing. In this case, the audio processing unit 134 may recognize an input having a change in size as a voice signal and filter it.

The noise processing unit 135 analyzes a noise waveform included in the voice signal and generates an interference wave that cancels the analyzed waveform. The audio signal separated by the audio processing unit 134 may contain a noise signal. In this case, after analyzing the waveform of the noise, if an interference wave opposite to the analyzed waveform is generated, the noise can be actively removed. Even in this case, the noise processing unit 135 may perform noise processing by correcting the interference wave so that the synthesized wave including the remaining noise is canceled again after receiving the feedback of the synthesized wave of the noise signal and the interference wave through the microphone 111. have.

The communication unit 136 communicates with the user terminal 200 and the interpretation and translation server 300. The communication unit 136 includes communication modules corresponding to various communication networks of the network 400, for example, a Bluetooth module, a WiFi module, an Ethernet, a USB module, and a cellular wireless communication module. In particular, a USB module as wired communication, a Bluetooth module, a Zigbee module, an NFC module, and a WiFi module as short-range communication may be included in the communication unit 136.

In addition, the PCB 130 includes a voice recognition DB, a translation DB, and a storage device that stores voice. A client program for linking the user terminal 200 and the interpretation and translation server 300 is stored in the storage device. Here, the storage device includes a volatile RAM, a nonvolatile ROM, and a flash memory, and stores various digital files according to their functions. In particular, by storing the TTS engine in the storage device, the speech synthesis function of the interpretation and translation server 300 may be performed directly in the smart clothing 100.

There is no guarantee that one side's utterance will end between the talkers and the other's utterance will be. Therefore, it is necessary to distinguish between different types of voices input at the same time in speech that can occur simultaneously. To solve this problem, the control unit 131 determines a characteristic of a tone using the sound spectrum of the speaker's voice, and filters the voice of a heterogeneous language simultaneously uttered using the determined feature of the tone using a filter. Accordingly, speakers of heterogeneous language voices are distinguished, and voices of heterogeneous languages may be separated from each other by filtering.

Further, the control unit 131 is characterized in that, with respect to the voices of the heterogeneous languages spoken at the same time, by using a scoring according to the translation result of the sample voices, it is characterized in that the different languages correspond to the languages of which countries.

Specifically, when the voices of English and Korean are mixed and input, one voice signal for the filtered voice signal according to the tone of the English speaker and the voice of the Korean speaker, and the other voice signal It attempts to translate into Korean and English, converts it into a score, and selects the translation with the highest score to determine which language is the corresponding language.

According to an embodiment of the present invention, language setting that the smart clothing 100 or the user terminal 200 will be in charge of may be automatically performed using a client program. That is, the control unit 131 may automatically set Korean as a starting language by referring to the language set in the user terminal 200.

Additionally, when the other user terminal 200 set as a different language is within a certain distance, a starting word and an object word may be automatically set by referring to the language setting of the other user terminal 200.

In addition, the control unit 131 scores the completeness of the translation using samples of speech spoken through the interpretation and interpretation unit 133, and automatically sets the language with the highest score as the destination language. For example, if the spoken voice of the other party corresponds to a foreign language other than Korean, the user terminal 200 translates the spoken voice into English, Japanese, Chinese, etc., and then scores the translated result and receives the highest score. ) Is set as the destination language for Korean, which is the starting language.

Hereinafter, an interpretation and translation method (S100) using smart clothing according to an embodiment of the present invention will be described.

5 is an exemplary diagram of an interpretation and translation method using smart clothing according to an embodiment of the present invention.

5, the interpretation and translation method (S100) using smart clothing according to an embodiment of the present invention includes steps S110 to S140.

First, the spoken voice is collected through the input unit 110 of the smart clothing 100 and converted into a voice signal. The input voice is converted into a voice signal through the microphone 111 of the input unit 110.

Next, the audio signal is preprocessed using at least one of the audio processing unit 134 and the noise processing unit 135 (S120).

The step of pre-processing the voice signal includes at least one of the following steps.

The voice signal is recognized by using the difference in the volume of the voice according to the location of the microphone and filtered (S121).

Analyzes the waveform of the noise signal included in the speech signal, generates an interference wave that cancels the analyzed waveform, and corrects the interference wave so that the synthesized wave is canceled by receiving feedback from the noise signal and the synthesized wave of the interference wave S122).

6 is an exemplary diagram of a preprocessing step of a voice signal according to an embodiment of the present invention.

Referring to FIG. 6, a voice including noise input through the microphone 111 is converted into a voice signal and a noise signal while passing through the microphone 111. The audio processing unit 134 recognizes voice signals other than noise by comparing the output of a plurality of microphones at different locations included in the smart clothing 100, for example, by comparing the output of the front microphone with the output of the rear microphone. And filter it.

The filtered voice signal still contains noise. The noise processing unit 135 generates a first-order interference wave capable of first-order destructive interference with a noise signal included in the voice signal. In addition, the primary interference wave is corrected by separating the noise still contained in the output voice signal and adding a secondary interference wave for secondary destructive interference to the separated noise to the primary interference wave.

Next, translation and interpretation are performed through a voice recognition process using the voice signal (S130).

Here, the translation and interpretation may be performed by the smart clothing 100 alone or through the user terminal 100 and the interpretation and translation server 300 communicating by wire or wirelessly.

Next, the translation and interpretation results are output in at least one form of text and synthesized voice using smart clothing (S140). Here, the smart clothing 100 may control the volume of the voice output through the output unit 120 and the size of the text using the sensing information of the distance sensor.

The interpretation and translation method (S100) using smart clothing according to an embodiment described through the above drawings may be implemented in the form of a recording medium including an instruction set executable by a computer such as a program module executed by a computer. I can. Computer-readable media can be any available media that can be accessed by a computer, and includes both volatile and nonvolatile media, removable and non-removable media. Further, the computer-readable medium may include both computer storage media and communication media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Communication media typically includes computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave, or other transmission mechanism, and includes any information delivery media.

As described above, according to an embodiment of the present invention, an interpretation and translation result may be output through audio and video using smart clothing.

In addition, it is possible to output an interpretation and translation result through a service of a remote interpretation/translation server using a communication function of a user terminal connected by wire or wirelessly.

In addition, it is possible to output the result of interpretation and translation so that the size of the voice and text output is automatically adjusted according to the distance between the talker or between the speaker and the audience using smart clothing.

In addition, external noise can be recognized and actively removed through smart clothing.

The above description of the present invention is for illustrative purposes only, and those of ordinary skill in the art to which the present invention pertains will be able to understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, it should be understood that the embodiments described above are illustrative and non-limiting in all respects. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as being distributed may also be implemented in a combined form.

The scope of the present invention is indicated by the claims to be described later rather than the detailed description, and all changes or modified forms derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention. do.

100: smart clothing 110: input unit
111: microphone 112: keyboard
120: output unit 121: display device
122: speaker 130: PCB
131: control unit 132: power supply unit
133: interpretation and translation unit 134: audio processing unit
135: noise processing unit 140: distance sensor
150: charging unit 200: user terminal
300: interpretation server 400: network
500: computing device

Claims (11)

In the interpretation and translation system using smart clothing that performs translation and interpretation for voice spoken through connection and communication between smart clothing and a user terminal and an interpretation and translation server,
The smart clothing,
An input unit for collecting speech for conversion from spoken speech to speech signal;
An output unit for outputting the translated content of the voice content recognized using the voice signal in at least one of text and synthesized voice;
A distance sensor for recognizing a distance from a user to a talker or an audience;
A control unit controlling a volume of a voice and a volume of a text output through the output unit by using sensing information of the distance sensor;
A power supply unit supplying power to the input unit and the output unit; And
A charging unit for charging the user terminal using the power unit; Including,
The interpretation and translation server,
A voice recognition module and a voice recognition DB for recognizing voices collected from the input unit;
Includes; a translation module and a translation DB for analyzing the speech of the talker uttered in the starting language recognized by the speech recognition module and translating the target language,
The speech recognition DB is a DB constructed by learning speech due to a plurality of speech examples using a deep learning algorithm and reducing or expanding the recognition range according to the frequency of speech content,
The translation DB is a DB constructed by learning a plurality of translation examples using a deep learning algorithm, expanding the colloquial expression according to the frequency of the translation examples, and reducing the written expression.
Interpretation and translation system using smart clothing. delete delete delete The method according to claim 1,
The smart clothing,
It includes a plurality of microphones of the input unit over the front and rear,
An interpretation and translation system using smart clothing, further comprising an audio processing unit for recognizing and filtering a voice signal by using a difference in volume of a voice according to a location of a microphone. The method according to claim 1,
The smart clothing,
Noise that analyzes the waveform of the noise signal included in the voice signal, generates an interference wave that cancels the analyzed waveform, and corrects the interference wave so that the synthesized wave is canceled by receiving feedback from the noise signal and the synthesized wave of the interference wave Interpretation and translation system using smart clothing, characterized in that it further comprises a processing unit. delete The method according to claim 1,
The output unit includes a display device,
The display device,
It is located on at least one of the front of the chest and abdomen and the back of the back of the smart clothing so that others other than the user can see it,
An interpretation and translation system using smart clothing, characterized in that it is located on the arm of the smart clothing for the user to see. delete delete delete

Images