KR20250024214A - Real-time translation subtitle providing system that enables real-time sharing of translated subtitle information - Google Patents

Abstract

The present invention relates to a real-time translation subtitle provision system capable of real-time sharing of translation subtitle information, and more specifically, to a real-time translation subtitle provision system capable of managing a speaker's voice information or voice information within a video in an international conference hall or real-time video streaming environment, converting it to STT and translating it into a requested language to provide subtitle information to a requesting user, and providing the subtitle information to another user at the user's request, so that the translation subtitle information can be shared in real time.
To this end, the present invention includes a management server which receives voice information of a speaker for whom subtitle provision request information has been received or voice information of video data, generates first subtitle information in the form of text through STT (Speech-To-Text) conversion for a first language, which is a language of the corresponding voice information, and translates the generated first subtitle information into a requested second language to generate second subtitle information in the form of text; a host terminal which is connected to the management server so as to enable data communication and transmits voice information of a specific speaker or voice information of specific video data to perform a subtitle provision request, receives the generated first subtitle information and second subtitle information from the management server, and selectively transmits subtitle sharing request information for the received first subtitle information and second subtitle information to the management server to receive shared Web connection information for receiving the first subtitle information and the second subtitle information from the management server; and a client terminal which receives the shared Web connection information from the host terminal and receives the first subtitle information and the second subtitle information.

Description

{Real-time translation subtitle providing system that enables real-time sharing of translated subtitle information}

The present invention relates to a real-time translation subtitle provision system capable of real-time sharing of translation subtitle information, and more specifically, to a real-time translation subtitle provision system capable of managing a speaker's voice information or voice information within a video in an international conference hall or real-time video streaming environment, converting it to STT and translating it into a requested language to provide subtitle information to a requesting user, and providing the subtitle information to another user at the user's request, so that the translation subtitle information can be shared in real time.

In general, people who are not good at foreign languages often experience discomfort and difficulty in communicating when the speakers are foreigners at international conferences or watching videos.

However, without the help of someone who knows the foreign language well, you have to look up dictionaries or phrasebooks on your own to communicate with foreigners, which makes it difficult to respond in real time and communicate properly.

Recently, with the advancement of smartphone technology, automatic translation machines have been developed, and their reliability has been put to practical use in the case of Korean and Japanese, which have similar language systems. In the case of English, the language systems are different from Korean, so the reliability of translation is lower than in the case of Japanese, but translation machines and their translation programs that can be practically used for short texts or simple complex sentences are widely known.

Meanwhile, much research effort is being put into voice recognition for machines to recognize human voices, and with the development of sampling technology and self-learning functions using neural networks, the automatic voice recognition rate is increasing and it is being put to practical use in some fields.

In addition, the technology of outputting text as speech is already widely used in voice output devices that create a database (DB) of pronunciation combinations of each letter, pronunciation of other words, pronunciation of sentences, etc. and output them as speech.

In addition, we provide interpretation services that convert foreign languages into the user's native language using the smart phone that modern people always carry with them.

However, the translation systems that are currently available in various forms are not suitable for 1:N conversations, and there is a problem that separate networks must be prepared in advance to provide translation information with a certain degree of smoothness.

Therefore, there is a need for improvement in the system that can simultaneously provide real-time translation of speech information of speakers speaking at international conferences or conferences, or of video data streaming live, to many users, as well as provide translation information quickly and easily to many users.

Korean Patent No. 10-1753649

The present invention has been made to solve the above problems, and the purpose of the present invention is to provide a real-time translation subtitle provision system in which a management server receives a speaker's voice information or a voice information within a video in an international conference room or real-time video streaming environment, performs STT conversion and translation into a requested language, and provides subtitle information to a requesting user, and also provides the subtitle information to another user at the request of the user, so that the translated subtitle information can be shared in real time.

To achieve the above purpose, the present invention has the following features.

The present invention comprises: a management server which receives voice information of a speaker for whom subtitle provision request information has been received or voice information of video data, generates first subtitle information in the form of text through STT (Speech-To-Text) conversion for a first language, which is the language of the corresponding voice information, and performs translation based on artificial intelligence learned on the generated first subtitle information into a requested second language to generate second subtitle information in the form of text; a host terminal which is connected to the management server so as to enable data communication and transmits voice information of a specific speaker or voice information of specific video data to perform a subtitle provision request, receives the first subtitle information and the second subtitle information generated from the management server, and selectively transmits subtitle sharing request information to the management server for the received first subtitle information and the second subtitle information to receive shared Web connection information for receiving the first subtitle information and the second subtitle information from the management server; and a client terminal which receives the shared Web connection information from the host terminal and is connected to the management server so as to enable data communication, and receives the first subtitle information and the second subtitle information from the management server.

Here, if the audio information in which the subtitle provision request information is received is video data, the host terminal overlays at least one of the first subtitle information and the second subtitle language received from the management server on the output screen of the video data and outputs it.

In addition, the shared Web connection information is URL (Uniform Resource Locator) information or QR code information of a Web page generated by the management server, and the management server transmits the first subtitle information and the second subtitle information to a client terminal that has accessed the Web page through the shared Web connection information, and includes a shared management unit that receives conversation information from the host terminal and at least one client terminal and outputs it on the Web page to enable two-way communication, wherein the shared management unit provides the conversation information received from each of the client terminals by translating it into a language set by the host terminal when providing it to the host terminal.

In addition, the host terminal further includes a host conversion unit that receives voice information of a speaker or voice information of video data, performs STT (Speech-To-Text) conversion on a first language, which is the language of the voice information, to generate first subtitle information in the form of text, and a host translation unit that performs translation based on learned artificial intelligence on the first subtitle information generated from the host conversion unit, but translates the translation into a second language set in advance, to generate second subtitle information in the form of text.

In addition, an interpreter terminal is further included, which is connected to the management server so as to enable data communication in response to an interpretation request from the host terminal, receives voice information of a specific speaker requested or voice information of video data and first subtitle information from the management server, and provides real-time interpretation voice information that performs interpretation in a second language set at the time of the interpretation request for the voice information and first subtitle information to the management server.

In addition, the management server performs STT conversion on real-time interpretation voice information received from the interpreter terminal to generate second subtitle information, and provides the generated second subtitle information together with the first subtitle information to the requested host terminal. In addition, when the subtitle sharing function is being performed, the management server provides the first subtitle information and the second subtitle information together to a client terminal connected through the shared Web connection information.

In addition, a stenographer terminal is further included, which is connected to the management server so as to enable data communication in response to a stenography request from the host terminal, receives voice information of a specific speaker requested or voice information of video data from the management server, and provides real-time stenography information in text form for the voice information to the management server; and the management server performs an artificial intelligence-based translation of the real-time stenography information received from the stenographer terminal to generate second subtitle information, and provides the real-time stenography information and the second subtitle information to the host terminal that has requested the stenography, and when a subtitle sharing function is being performed, provides the real-time stenography information and the second subtitle information together to a client terminal connected via the shared Web connection information.

According to the present invention, translation information for a speaker's voice information or voice information of real-time streaming video data can be provided in real time, and, in response to a sharing request from a host terminal, a large number of other users can quickly and easily receive real-time translation information without a separate application installation process or login process.

FIG. 1 is a drawing showing a schematic configuration of a real-time translation subtitle provision system according to one embodiment of the present invention.
FIG. 2 is a drawing showing the internal configuration of a real-time translation subtitle provision system according to one embodiment of the present invention.
FIG. 3 is a drawing showing the internal configuration of a host terminal according to another embodiment of the present invention.
FIG. 4 is a drawing showing a main display screen in a dedicated application of a host terminal according to one embodiment of the present invention.
FIG. 5 is a drawing showing a state in which first subtitle information or second subtitle information is overlaid on video data and output in a host terminal according to one embodiment of the present invention.
FIGS. 6 and 7 are drawings showing a subtitle sharing request screen in a dedicated application of a host terminal according to one embodiment of the present invention.
FIG. 8 is a drawing showing a web page display screen of a client terminal according to one embodiment of the present invention.
FIG. 9 is a diagram showing the internal configuration of a real-time translation subtitle provision system according to another embodiment of the present invention.
FIG. 10 is a drawing showing the internal configuration of a real-time translation subtitle providing system according to another embodiment of the present invention.
FIG. 11 is a drawing showing the internal configuration of a real-time translation subtitle providing system according to another embodiment of the present invention.

In order to explain the present invention, the operational advantages of the present invention, and the purpose achieved by practicing the present invention, preferred embodiments of the present invention will be exemplified and examined with reference thereto.

First, the terminology used in this application is only used to describe specific embodiments, and is not intended to limit the present invention, and the singular expression may include plural expressions unless the context clearly indicates otherwise. Also, in this application, it should be understood that the terms "comprise" or "have" and the like are intended to specify the presence of a feature, number, step, operation, component, part or combination thereof described in the specification, but do not exclude in advance the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts or combinations thereof.

In describing the present invention, if it is determined that a detailed description of a related known configuration or function may obscure the gist of the present invention, the detailed description will be omitted.

FIG. 1 is a drawing showing a schematic configuration of a real-time translation subtitle providing system according to an embodiment of the present invention, FIG. 2 is a drawing showing the internal configuration of a real-time translation subtitle providing system according to an embodiment of the present invention, FIG. 3 is a drawing showing the internal configuration of a host terminal according to another embodiment of the present invention, and FIG. 4 is a drawing showing a main display screen in a dedicated application of a host terminal according to an embodiment of the present invention.

In addition, FIG. 5 is a drawing showing a state in which first subtitle information or second subtitle information is overlaid on video data and output in a host terminal according to an embodiment of the present invention, FIGS. 6 and 7 are drawings showing a subtitle sharing request screen in a dedicated application of a host terminal according to an embodiment of the present invention, and FIG. 8 is a drawing showing a web page display screen of a client terminal according to an embodiment of the present invention.

Referring to the drawings, a real-time interpretation service system (1000) according to an embodiment of the present invention comprises: a management server (100) that receives voice information of a speaker for whom subtitle provision request information has been received or voice information of video data, generates first subtitle information in the form of text through STT (Speech-To-Text) conversion for a first language, which is the language of the corresponding voice information, and performs translation based on artificial intelligence learned on the generated first subtitle information into the requested second language to generate second subtitle information in the form of text; a host terminal (200) that is connected to the management server (100) so as to enable data communication and transmits voice information of a specific speaker or voice information of specific video data to perform a subtitle provision request, receives the first subtitle information and the second subtitle information generated from the management server (100), selectively transmits subtitle sharing request information for the received first subtitle information and the second subtitle information to the management server (100), and receives shared Web connection information for receiving the first subtitle information and the second subtitle information from the management server; and the host terminal (200) It is composed of a client terminal (300) that receives the shared Web connection information from the terminal (200), connects to the management server (100) so that data communication is possible, and receives the first subtitle information and the second subtitle information from the management server (100).

Here, the management server (100) receives voice information of a specific speaker or voice information of video data requested by the host terminal (200), performs STT (Speech-To-Text) conversion of the voice information in real time, generates first subtitle information in text form as a result of the conversion, and generates second subtitle information in text form with an artificial intelligence-based translation performed based on the generated first subtitle information.

Accordingly, the first subtitle information and second subtitle information generated are transmitted in real time to the host terminal (200) that requests subtitle provision.

To this end, the management server (100) is configured with a server communication unit (110) that enables data communication with the host terminal (200) and the client terminal (300), and a subtitle management unit (120) that generates the first subtitle information and the second subtitle information in real time based on voice information of a specific speaker or voice information of video data transmitted by the host terminal (200) when there is a request for subtitle provision from the host terminal (200) through the server communication unit (310).

Here, the subtitle management unit (120) is configured to include an STT conversion unit (121) that performs STT (Speech-To-Text) conversion into a first language, which is the language of the corresponding voice information for which the host terminal (200) has requested subtitle provision, to generate first subtitle information in text form, and an artificial intelligence translation unit (122) that performs artificial intelligence-based translation based on the generated first subtitle information to generate second subtitle information in text form.

Of course, the subtitle management unit (120) transmits the first subtitle information and the second subtitle information generated from the STT conversion unit (121) and the artificial intelligence translation unit (122) in real time to the host terminal (200) that requested the subtitle provision through the server communication unit (110).

In addition, the management server (100) includes a sharing management unit (130) that, when the host terminal (200) makes a subtitle sharing request to the server communication unit (110), creates a web page for subtitle sharing, transmits shared web connection information to the host terminal (200) to access the web page, and transmits the first subtitle information and the second subtitle information to at least one client terminal (300) accessing the web page.

At this time, it is desirable that the shared web connection information be URL (Uniform Resource Locator) information or QR code information for accessing the relevant web page.

In addition, the shared management unit (130) may be configured to provide a chat function between a user of the host terminal (200) and a user of the client terminal (300) by receiving conversation information from each of the host terminal (200) and the client terminal (300) and outputting it on the web page.

At this time, it is preferable that the shared management unit (130) be configured to provide the conversation information received from each client terminal (300) to the host terminal (200) by translating it into a language set by the host terminal (200).

In addition, the shared management unit (130) may further include a TTS (Text-to-Speech) conversion unit (not shown in the drawing) that outputs voice information through voice synthesis based on second subtitle information translated into a second language set in advance by each user at the request of the host terminal (200) or client terminal (300).

Accordingly, the shared management unit (130) of the management server (100) that receives a voice output request from the host terminal (200) or client terminal (300) provides all of the first subtitle information, the second subtitle information, and the second subtitle voice information obtained by performing voice synthesis on the second subtitle information to the host terminal (200) or client terminal (300).

In addition, the STT conversion unit (121) and the artificial intelligence translation unit (122) of the management server (100) can be configured to determine the accuracy of the first subtitle information and the second subtitle information they each generate, for each entire phrase or sentence.

Accordingly, based on the accuracy thereof, it is possible to configure the first subtitle information and the second subtitle information not to be generated for a specific phrase or sentence, or to perform separate distinguishing processing on the generated first subtitle information and the second subtitle information and transmit them to the host terminal (200) or the client terminal (300).

That is, if the accuracy is determined to be too low based on a pre-set criterion, the first subtitle information and the second subtitle information are not generated, and if the accuracy is determined to be neither too low nor too high based on a pre-set criterion, the first subtitle information and the second subtitle information are generated, but the first subtitle information and the second subtitle information are separately processed to enable a user of a host terminal (200) or client terminal (300) receiving the first subtitle information to recognize that the phrase is low in accuracy.

In addition, the STT conversion unit (121) and the artificial intelligence translation unit (122) can be configured to generate multiple first subtitle information and second subtitle information for each phrase or each sentence during STT conversion or translation.

This is because the STT conversion unit (121) and the artificial intelligence translation unit (122) can generate first subtitle information or second subtitle information as one phrase or one sentence when recognizing or translating food, and in some cases, multiple phrases or sentences that can be voice-recognized or translated can be generated.

In this case, the STT conversion unit (121) and the artificial intelligence translation unit (122) determine the priority based on accuracy, specify the corresponding phrase or sentence in order of high priority, and send the corresponding phrase or sentence with the highest priority to the host terminal (200) or client terminal (300), but transmit it after separate processing.

Accordingly, when the host terminal (200) or the client terminal (300) receives the first subtitle information or the second subtitle information that has undergone the corresponding distinction processing, it can request the management server (100) for the first subtitle information or the second subtitle information for the corresponding phrase or sentence with a lower priority, and if there is a request from the host terminal (200) or the client terminal (300), the STT conversion unit (121) and the artificial intelligence translation unit (122) specify the first subtitle information and the second subtitle information, excluding the highest priority for the corresponding phrase or sentence, by priority and transmit them to the host terminal (200) or the client terminal (300).

In addition, it may be configured to receive user selection information for the corresponding phrase or sentence from the host terminal (200) or client terminal (300) that received it, and such selection information is information that the user of the host terminal (200) or client terminal (300) judges that the first subtitle information or the second subtitle information of a lower rank than the first subtitle information or the second subtitle information of the highest rank judged by the STT conversion unit (121) and the artificial intelligence translation unit (122) is more accurate in voice recognition or translation.

When the STT conversion unit (121) and artificial intelligence translation unit (122) receive this data, a higher quality artificial intelligence model can be created by utilizing this data as learning data.

Meanwhile, the host terminal (200) is equipped to perform a subtitle provision request by transmitting voice information of a specific speaker or voice information of specific video data to the management server (100), and to receive the first subtitle information and second subtitle information generated from the management server (100).

The user of such a host terminal (200) may be a speaker giving a lecture or presentation, or may be another person.

A host terminal (200) of this type is configured to include a host communication unit (210) that enables data communication with the management server (100) and the client terminal (300), a user input unit (220) that receives input information from a user, a screen management unit (230) that provides output information to the user, and a host control unit (240) that, when a subtitle provision request is received from the user input unit (220), transmits voice information of a specific speaker or voice information of specific video data to the management server (100) through the host communication unit (210), receives first subtitle information and second subtitle information from the management server (100), and provides them to the user through the screen management unit (230).

Here, when the host control unit (240) receives a subtitle sharing request from the user input unit (220) as shown in FIGS. 6 and 7, it transmits subtitle sharing request information to the management server (100) through the host communication unit (210) and receives shared Web connection information from the management server (100).

Accordingly, it is configured to transmit the shared Web connection information requested from the user input unit (220) to the client terminal (300) through the host communication unit (210).

In addition, the screen management unit (230) is configured to include a first subtitle display unit (231) in which first subtitle information is displayed, a second subtitle display unit (232) in which second subtitle information is displayed, a language setting unit (233) for setting the first language and the second language, and a subtitle sharing setting unit (234) for setting a subtitle sharing request, as shown in FIG. 4.

In addition, as shown in FIG. 5, when the voice information for which the subtitle provision request information is received is video data and overlay selection information is received from the user input unit (220), the screen management unit (230) can output at least one of the first subtitle information and the second subtitle language received from the management server (100) by overlaying it on the output screen of the video data.

In addition, according to an example of the present invention, the host terminal (200) may further include a host conversion unit (250) that receives voice information of a speaker or voice information of video data, performs STT (Speech-To-Text) conversion on a first language, which is the language of the voice information, to generate first subtitle information in the form of text, and a host translation unit (260) that performs translation based on learned artificial intelligence on the first subtitle information generated from the host conversion unit (250), but performs translation into a second language set in advance, to generate second subtitle information in the form of text.

In cases where data communication with the management server (100) is difficult due to poor communication environment, the host conversion unit (250) and the host translation unit (260) perform STT conversion by directly recognizing voice information of a speaker or video data in the host terminal (200), and performs artificial intelligence-based translation based on this, thereby generating first subtitle information and second subtitle information.

Meanwhile, the client terminal (300) receives the shared web connection information from the host terminal (200), connects to the management server (100) so that data communication is possible, and is provided with the first subtitle information and the second subtitle information from the management server (100).

The client terminal (300) is configured to include a client communication unit (310) that enables data communication with the management server (100) and the host terminal (200), a user input unit (320) that receives input information from a user, a screen management unit (330) that provides output information to the user, and a client control unit (340) that receives the shared Web connection information from the host terminal (200), and when connection selection information of the shared Web connection information is transmitted from the user input unit (320), connects to the management server (100) through a corresponding web page, receives first subtitle information and second subtitle information from the management server (100), and provides them to the user through the screen management unit (330).

Accordingly, when the client control unit (340) receives shared Web connection information from the host terminal (200), and accesses a web page corresponding to the shared Web connection information, it is connected to the management server (100) as shown in FIG. 8, and the first subtitle information and the second subtitle information are provided to the user by the screen management unit (330).

As illustrated in FIG. 8, a user of a client terminal (300) may be provided with a chat function with another client terminal user or a host terminal user, and may be provided with a function for setting the simultaneous output of first subtitle information and second subtitle information or displaying only second subtitle information.

In this way, the real-time translation subtitle provision system according to one embodiment of the present invention enables subtitle sharing settings on the host terminal (200), so that multiple other client terminals (300) can quickly and easily receive first subtitle information and second subtitle information on voice information of a specific speaker or voice information of real-time streaming video data without a separate login procedure or installation process of a dedicated application.

FIG. 9 is a diagram showing the internal configuration of a real-time translation subtitle provision system according to another embodiment of the present invention.

Referring to the drawings, the real-time translation subtitle provision system (1000) according to the present embodiment further includes an interpreter terminal (400) that is connected to the management server (100) so as to enable data communication in response to an interpretation request from a host terminal (200), receives voice information of a specific speaker requested or voice information of video data and first subtitle information from the management server (100), and provides real-time interpretation voice information that performs interpretation in a second language set at the time of the interpretation request for the corresponding voice information and first subtitle information to the management server (100).

Accordingly, the management server (100) performs STT conversion on real-time interpretation voice information received from the interpretation terminal (400) to generate second subtitle information, and provides the generated second subtitle information together with the first subtitle information to the host terminal (100) that requested the interpretation service.

In addition, as described in the real-time translation subtitle provision system (1000) described above, when the subtitle sharing function is being performed at the request of the current host terminal (200), the management server (100) provides the first subtitle information and the second subtitle information together to the client terminal (300) connected through the shared Web connection information.

FIG. 10 is a drawing showing the internal configuration of a real-time translation subtitle providing system according to another embodiment of the present invention.

Referring to the drawings, the real-time translation subtitle provision system (1000) according to the present embodiment further includes a stenographer terminal (500) that is connected to the management server (100) so as to enable data communication in response to a stenography request from the host terminal (200), receives voice information of a specific speaker requested or voice information of video data from the management server (100), and provides real-time stenography information in text form for the corresponding voice information to the management server (100).

Accordingly, the management server (100) performs an artificial intelligence-based translation of real-time stenography information received from the stenographer terminal (500) to generate second subtitle information, and provides the real-time stenography information and second subtitle information to the host terminal (200) that requested stenography.

In addition, as described in the real-time translation subtitle provision system (1000) described above, if the subtitle sharing function is being performed at the request of the current host terminal (200), the management server (100) can provide the real-time stenography information and the second subtitle information together to the client terminal (300) connected through the shared Web connection information.

In addition, the interpreter terminal (400) of the above-described embodiment may be further included in this embodiment so that when an interpretation request and a stenography request from the host terminal (200) are made at the same time, interpretation through the interpreter terminal (400) and stenography through the stenographer terminal (400) may be made at the same time.

FIG. 11 is a drawing showing the internal configuration of a real-time translation subtitle providing system according to another embodiment of the present invention.

Referring to the drawings, the real-time translation subtitle provision system (1000) according to the present embodiment may further include an inspector terminal (600) that inspects and corrects the result information, i.e., the first subtitle information or the second subtitle information, output from at least one of the STT conversion unit (121) and the artificial intelligence translation unit (122) of the management server (100) upon request from the host terminal (200) or according to preset settings, and transmits the correction information to the management server (100).

At this time, the STT conversion unit (121) determines whether the result information generated during voice recognition is suitable as set in advance. If the suitableness does not exceed the threshold value of the first suitable criterion set in advance, it determines that voice recognition is impossible and does not generate the first subtitle information for the corresponding phrase or sentence.

In addition, if the accuracy of voice recognition is judged to be low even if the first suitability criterion is exceeded but the threshold of the second suitability criterion set in advance is not exceeded, the phrase or sentence in question may be configured to be processed separately and the first subtitle information transmitted.

Accordingly, if there is a phrase or sentence that is not generated as first subtitle information because it does not exceed the threshold value of the first suitability criterion in the STT conversion unit (121), or if there is a phrase or sentence that is generated as first subtitle information but is processed separately because it does not exceed the threshold value of the second suitability criterion, the inspector terminal (600) generates correction information for this and transmits it to the management server (100) in real time.

Accordingly, when the management server (100) receives correction information from the inspector terminal (600), it is configured to transmit the correction information to the first subtitle information already provided to the host terminal (200) or client terminal (300) to replace the first subtitle information for the existing phrase or sentence.

Although the present invention has been described with reference to one embodiment shown in the drawings, this is merely exemplary, and those skilled in the art will understand that various modifications and equivalent other embodiments are possible therefrom.

Therefore, the true technical protection scope of the present invention should be determined by the technical idea of the appended patent claims.

100 : Management Server
110: Server communication section 120: Subtitle management section
121: STT conversion unit 122: AI translation unit
130 : Shared Management Department
200 : Host terminal
210: Host communication section 220: User input section
230: Screen management unit 231: First subtitle display unit
232: Second subtitle display section 233: Language setting section
234: Subtitle sharing settings section 240: Host control section
250: Host conversion section 260: Host translation section
300 : Client terminal
310: Client communication section 320: User input section
330: Screen management unit 340: Client control unit
400 : Interpreter Terminal
500 : Stenographer's Terminal
1000: Real-time translation subtitle provision system

Claims (7)

A management server that receives voice information of a speaker or voice information of video data from which subtitle provision request information has been received, generates first subtitle information in the form of text through STT (Speech-To-Text) conversion for a first language, which is the language of the corresponding voice information, and performs translation based on artificial intelligence learned on the generated first subtitle information into the requested second language to generate second subtitle information in the form of text;
A host terminal connected to the management server so as to enable data communication, transmits voice information of a specific speaker or voice information of specific video data to perform a subtitle provision request, receives the first subtitle information and the second subtitle information generated from the management server, and selectively transmits subtitle sharing request information for the received first subtitle information and the second subtitle information to the management server to receive shared Web connection information for providing the first subtitle information and the second subtitle information from the management server; and
A client terminal that receives the shared Web connection information from the host terminal and is connected to the management server so as to enable data communication, and receives the first subtitle information and the second subtitle information from the management server;
Real-time translation subtitle provision system.
In the first paragraph,
The above host terminal is,
If the received audio information of the subtitle provision request information is video data, at least one of the first subtitle information and the second subtitle language received from the management server is overlaid on the output screen of the video data and outputted.
Real-time translation subtitle provision system.
In the first paragraph,
The above shared web connection information is,
The URL (Uniform Resource Locator) information or QR code information of the web page generated by the above management server,
The above management server
A shared management unit is included that transmits the first subtitle information and the second subtitle information to a client terminal that has accessed a web page through the shared web connection information, and receives conversation information from the host terminal and at least one client terminal and outputs it on the web page, thereby enabling two-way communication.
The above shared management department
When providing conversation information received from each client terminal to the host terminal, translation is performed and provided in a language set by the host terminal.
Real-time translation subtitle provision system.
In the first paragraph,
The above host terminal is
A host conversion unit that receives voice information of a speaker or voice information of video data, performs STT (Speech-To-Text) conversion on a first language, which is the language of the voice information, to generate first subtitle information in the form of text, and a host translation unit that performs translation based on learned artificial intelligence on the first subtitle information generated from the host conversion unit, but performs translation into a second language set in advance to generate second subtitle information in the form of text.
Real-time translation subtitle provision system.
In the first paragraph,
An interpreter terminal is further included, which is connected to the management server so as to enable data communication in response to an interpretation request from the host terminal, receives voice information of a specific speaker requested or voice information of video data and first subtitle information from the management server, and provides real-time interpretation voice information that interprets the voice information and first subtitle information in a second language set at the time of the interpretation request to the management server.
Real-time translation subtitle provision system.
In paragraph 5,
The above management server,
The real-time interpretation voice information received from the above interpreter terminal is converted into STT to generate second subtitle information, and the generated second subtitle information is provided to the requested host terminal together with the first subtitle information, and when the subtitle sharing function is being performed, the first subtitle information and the second subtitle information are provided together to the client terminal connected through the shared Web connection information.
Real-time translation subtitle provision system.
In the first paragraph,
A stenographer terminal is further included, which is connected to the management server so as to enable data communication in response to a stenography request from the host terminal, receives voice information of a specific speaker or voice information of video data requested from the management server, and provides real-time stenography information in text form for the voice information to the management server;
The above management server,
Real-time stenography information received from the above stenographer terminal is translated based on artificial intelligence to generate second subtitle information, and the real-time stenography information and second subtitle information are provided to the host terminal that requested the stenography. In the case where the subtitle sharing function is being performed, the real-time stenography information and second subtitle information are provided together to the client terminal connected through the shared Web connection information.
Real-time translation subtitle provision system.