CN112818704B - Multilingual translation system and method based on inter-thread consensus feedback - Google Patents
Multilingual translation system and method based on inter-thread consensus feedback Download PDFInfo
- Publication number
- CN112818704B CN112818704B CN202110034160.8A CN202110034160A CN112818704B CN 112818704 B CN112818704 B CN 112818704B CN 202110034160 A CN202110034160 A CN 202110034160A CN 112818704 B CN112818704 B CN 112818704B
- Authority
- CN
- China
- Prior art keywords
- translation
- voice
- thread
- translated
- language
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000013519 translation Methods 0.000 title claims abstract description 174
- 238000000034 method Methods 0.000 title claims abstract description 21
- 238000012544 monitoring process Methods 0.000 claims description 29
- 238000004891 communication Methods 0.000 claims description 10
- 230000005540 biological transmission Effects 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000004888 barrier function Effects 0.000 description 2
- 241000282414 Homo sapiens Species 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/40—Processing or translation of natural language
- G06F40/55—Rule-based translation
- G06F40/56—Natural language generation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/40—Processing or translation of natural language
- G06F40/58—Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation
Abstract
The invention provides a multilingual translation system and a multilingual translation method based on inter-thread consensus feedback. The translation system comprises a first number of first translation engines, a second number of second translation engines and a voice input module to be translated; the to-be-translated voice input module receives a voice input signal sent by a user terminal; the first number of first translation engines comprises a plurality of first translation engines for translating the speech to be translated into target language speech; the second number of second translation engines comprises a plurality of second translation engines which can translate the target language voice into language texts corresponding to the voice to be translated; the second translation engine starts a plurality of threads when running, and each thread corresponds to one second translation engine; the translation system determines whether to generate at least one digest of the speech to be translated based on consensus feedback among the plurality of threads. The invention also provides a translation method realized based on the system.
Description
Technical Field
The invention belongs to the technical field of intelligent speech translation, and particularly relates to a multilingual translation system and method based on inter-thread consensus feedback.
Background
From ancient times to date, language translators can play an important role in solving the problems of communication gaps and barriers among people with different native language backgrounds. But this service is available to only a few people due to personnel cost limitations. For example, international organizations such as the united nations pay free translators more than $600 per day. The manual translation service is enjoyable by only a small number of people.
Over the past decade, with the rapid development of computer technology, speech technology and language processing technology, spoken language systems have advanced significantly, and some systems have emerged that are successful in a limited field, particularly as represented by the most important two spoken language systems, the spoken language translation system and the human-machine spoken language dialog system.
The voice real-time translation technology is applied to daily communication situations, and can help people with different native language backgrounds to carry out language communication. For example, the software for real-time translation of voice called Skype Translator, which is introduced by microsoft, can break the communication barrier between people, eliminate the solitary feeling caused by the inability of people to communicate with the external environment, and help people to talk easily across languages.
A speech recognition system as proposed in international application PCT/JP2018/040422 is used for recognizing speech comprising overlapping speech of a plurality of speakers. The system includes a hardware processor. The computer storage memory stores data and has computer-executable instructions stored thereon that, when executed by the processor, will implement the stored speech recognition network. The input interface receives an acoustic signal, the received acoustic signal comprising a mixture of speech signals of a plurality of speakers, wherein the plurality of speakers comprises a target speaker. The stored encoder and decoder networks of the speech recognition network are trained to transform the received acoustic signals into text for each target speaker. Such that the encoder network outputs a set of recognition codes and the decoder network uses the set of recognition codes to output text for each target speaker. The output interface transmits text for each target speaker.
However, in certain scenarios, such as multilingual voice conferencing, current machine translation models are difficult to compare with human beings. Currently, more than five thousand languages are used by people worldwide, however, only ten of these five thousand languages are widely used, which are: chinese, english, russian, spanish, indian, arabic, portuguese, bangla, german, and japanese. However, other languages have a small number of people, so that it is difficult to collect their voice data, and the corresponding voice data labeling work is quite difficult to implement, so that the accuracy of the translation work is difficult to be ensured. However, the important conference itself in multilingual voice conference situations requires the recording of accurate voice texts, while avoiding ambiguity and even errors in real-time machine spoken language translation. The prior art does not give an effective solution for this.
Disclosure of Invention
In order to solve the technical problems, the invention provides a multilingual translation system and a multilingual translation method based on inter-thread consensus feedback. The translation system comprises a first number of first translation engines, a second number of second translation engines and a voice input module to be translated; the to-be-translated voice input module receives a voice input signal sent by a user terminal; the first number of first translation engines comprises a plurality of first translation engines for translating the speech to be translated into target language speech; the second number of second translation engines comprises a plurality of second translation engines which can translate the target language voice into language texts corresponding to the voice to be translated; the second translation engine starts a plurality of threads when running, and each thread corresponds to one second translation engine; the translation system determines whether to generate at least one digest of the speech to be translated based on consensus feedback among the plurality of threads.
The invention also provides a translation method realized based on the system.
Specifically, in a first aspect of the present invention, there is provided a multi-lingual translation system based on inter-thread consensus feedback, the translation system comprising a first number of first translation engines, a second number of second translation engines, and a speech input module to be translated;
the to-be-translated voice input module is in remote wireless communication with a plurality of user terminals in different geographic positions and is used for receiving voice input signals sent by the user terminals;
the first number of first translation engines comprises a plurality of first translation engines for translating the speech to be translated into target language speech;
the second number of second translation engines comprises a plurality of second translation engines which can translate the target language voice into language texts corresponding to the voice to be translated;
the second translation engine starts a plurality of threads when running, and each thread corresponds to one second translation engine;
the translation system determines whether to generate at least one abstract of the voice to be translated based on consensus feedback among the plurality of threads;
the abstract comprises the user terminal number, the sending time and the voice input signal corresponding to the voice to be translated.
In the above technical solution, as an important component, in a specific implementation, the first translation result is input to a plurality of second translation engines, and the plurality of second translation engines output a plurality of second translation results.
Based on the above, the translation system determines whether to generate at least one abstract of the speech to be translated based on consensus feedback among the plurality of threads, and specifically comprises the following steps:
performing similarity comparison on each second translation result and the voice to be translated to obtain a plurality of similarity comparison values;
and if each similarity comparison value is lower than a preset threshold value, generating at least one abstract of the voice to be translated.
In a second aspect of the invention, a method for multilingual translation based on inter-thread consensus feedback is provided, the method being implemented based on a multi-threaded implementation.
Specifically, the method comprises the following steps:
s810: starting a monitoring thread, wherein the monitoring thread judges whether voice access exists or not; if so, entering the next step;
s820: starting an identification thread, wherein the identification thread is used for identifying the language category of the voice accessed by the monitoring thread;
s830: starting a grouping thread, wherein the grouping thread is used for grouping the voice accessed by the monitoring thread;
s840: starting a plurality of first translation threads, wherein the plurality of first translation threads correspond to the language types of the voices accessed by the monitoring threads;
s850: starting a plurality of second translation threads, wherein the plurality of second translation threads are communicated with the plurality of first translation threads through a data pipeline, and obtaining translation results output by the first translation threads;
s860: starting a consensus thread, wherein the consensus thread performs consensus calculation based on the output results of the plurality of second translation threads and the voice accessed by the monitoring thread;
s870: judging whether a summary recording thread needs to be started or not based on the result of the consensus calculation; if yes, entering the next step; otherwise, returning to step S810;
s880: and starting a digest recording thread, wherein the digest recording thread generates at least one digest of the voice accessed by the monitoring thread and records the digest into a system file.
The first translation thread comprises a plurality of first translation engines for translating the voice accessed by the monitoring thread into target language voice;
the second translation thread comprises a plurality of second translation engines which can translate the target language voice into the voice corresponding language text accessed by the monitoring thread.
By the technical scheme of the invention, various technical problems mentioned in the background art are solved.
Further advantages of the invention will be further elaborated in the description section of the embodiments in connection with the drawings.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a diagram of a multi-lingual translation system based on cross-thread consensus feedback according to an embodiment of the present invention
FIG. 2 is a schematic diagram of a transmission flow of a portion of a data stream in the system of FIG. 1
FIG. 3 is a control flow diagram of data transmission in the system of FIG. 1
FIG. 4 is a schematic diagram of threads involved in a multi-lingual translation method based on cross-thread consensus feedback implemented based on the system of FIG. 1
FIG. 5 is a main flow chart of a multi-lingual translation method based on cross-thread consensus feedback implemented based on the system of FIG. 1
Detailed Description
The invention will be further described with reference to the drawings and detailed description.
Referring to FIG. 1, a diagram of a multi-lingual translation system based on inter-thread consensus feedback according to an embodiment of the present invention is shown.
In fig. 1, the translation system includes a first number of first translation engines, a second number of second translation engines, and a speech input module to be translated.
More specifically, a first number of first translation engines forms a first translation engine group and a second number of second translation engines forms a second translation engine group.
The translation system further comprises a consensus module and a summary generation module.
The to-be-translated voice input module is in remote wireless communication with a plurality of user terminals in different geographic positions and is used for receiving voice input signals sent by the user terminals;
the first number of first translation engines comprises a plurality of first translation engines for translating the speech to be translated into target language speech;
the second number of second translation engines comprises a plurality of second translation engines which can translate the target language voice into language texts corresponding to the voice to be translated;
the second translation engine starts a plurality of threads when running, and each thread corresponds to one second translation engine;
the translation system determines whether to generate at least one abstract of the voice to be translated based on consensus feedback among the plurality of threads;
the abstract comprises the user terminal number, the sending time and the voice input signal corresponding to the voice to be translated.
The consensus feedback is completed by the consensus module, and the summary generation module is completed by the summary generation module.
See fig. 2-3 based on fig. 1.
As mentioned above, the to-be-translated voice input module is in remote wireless communication with a plurality of user terminals located at different geographic locations, and is configured to receive a voice input signal sent by the user terminal.
As an example as described in fig. 2, the user terminal may be a mobile terminal.
The user terminal receives a voice input signal input by a user through a voice input interface, embeds geographic position information of the user terminal in the voice input signal, and sends the voice input signal to the voice input module to be translated as voice to be translated.
In fig. 3, the mobile terminal collects an original voice signal, synthesizes the original voice signal with geographic position information, and generates a voice signal to be interpreted;
the to-be-translated voice input module comprises a voice preprocessing module, wherein the voice preprocessing module comprises a language recognition module and an input voice grouping module;
the language identification module identifies the language type of the current voice to be translated based on the geographic position information embedded in the voice to be translated.
At this time, the language type recognition module may directly determine the language type of the language to be translated based on the preset position information-language correspondence database if the language to be translated already contains at least the position information.
This way of judgment and recognition is particularly applicable to remote video/voice conferences in special circumstances, i.e. where the participants are located locally in each case, participating in the conference via a remote voice system.
Unlike the prior art, which needs to adopt complicated speech recognition and other techniques to determine the input language, the embodiment can avoid speech translation recognition and delay caused by the complicated recognition.
On this basis, the input voice grouping module groups the voice to be translated based on the language type of the voice to be translated, so as to obtain the grouping voice signal to be translated shown in fig. 3, wherein the language type of the language to be translated in each grouping is the same.
The first number is greater than the grouping number;
the input voice packet module comprises a plurality of packet data transmission channels, each packet data transmission channel being in unidirectional communication with a first translation engine;
the input voice grouping module sends the voice to be translated in each grouping to the corresponding first translation engine through a grouping data transmission channel.
The second number is less than or equal to the first number;
after the voice to be translated is input to the first translation engine, the first translation engine outputs a first translation result, namely a first grouping translation result shown in fig. 3;
and inputting the first translation result to a plurality of second translation engines, wherein the plurality of second translation engines output a plurality of second translation results, namely the second packet translation results shown in fig. 3.
The translation system determines whether to generate at least one abstract of the voice to be translated based on the common knowledge feedback among the threads, and specifically comprises the following steps:
performing similarity comparison on each second translation result and the voice to be translated to obtain a plurality of similarity comparison values;
and if each similarity comparison value is lower than a preset threshold value, generating at least one abstract of the voice to be translated.
The translation system also comprises a voice text recognition system;
the voice text recognition system translates the original voice signal into an original voice text;
the second translation result includes the plurality of second translation texts;
and comparing the similarity between the original voice text and the plurality of second translation texts to obtain a plurality of similarity comparison values.
In a specific implementation, the first translation result is input to a plurality of second translation engines, and the plurality of second translation engines output a plurality of second translation results;
and, the first translation engine translates the grouped to-be-translated speech signals into target language speech (first translation result); the second translation engine can translate the first translation result into language texts corresponding to the voice signals to be translated.
See fig. 4 and 5 based on fig. 1-3.
Fig. 4 shows a multilingual translation method based on inter-thread consensus feedback, which is implemented based on a plurality of threads, including a snoop thread, an identification thread, a grouping thread, a first translation thread, a second translation thread, a consensus thread, and a summary record thread.
FIG. 5 presents a specific thread start sequence in a method flow manner:
the method shown in fig. 5 includes steps S810-S880, where the specific corresponding thread starting manner of each step is as follows:
s810: starting a monitoring thread, wherein the monitoring thread judges whether voice access exists or not; if so, entering the next step;
s820: starting an identification thread, wherein the identification thread is used for identifying the language category of the voice accessed by the monitoring thread;
s830: starting a grouping thread, wherein the grouping thread is used for grouping the voice accessed by the monitoring thread;
s840: starting a plurality of first translation threads, wherein the plurality of first translation threads correspond to the language types of the voices accessed by the monitoring threads;
s850: starting a plurality of second translation threads, wherein the plurality of second translation threads are communicated with the plurality of first translation threads through a data pipeline, and obtaining translation results output by the first translation threads;
s860: starting a consensus thread, wherein the consensus thread performs consensus calculation based on the output results of the plurality of second translation threads and the voice accessed by the monitoring thread;
s870: judging whether a summary recording thread needs to be started or not based on the result of the consensus calculation;
if yes, entering the next step; otherwise, returning to step S810;
s880: and starting a digest recording thread, wherein the digest recording thread generates at least one digest of the voice accessed by the monitoring thread and records the digest into a system file.
The first translation thread comprises a plurality of first translation engines for translating the voice accessed by the monitoring thread into target language voice;
the second translation thread comprises a plurality of second translation engines which can translate the target language voice into the voice corresponding language text accessed by the monitoring thread.
The abstract comprises a user terminal number, a sending time and a voice input signal which correspond to the voice accessed by the monitoring thread.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (9)
1. A multi-lingual translation system based on inter-thread consensus feedback, the translation system comprising a first number of first translation engines, a second number of second translation engines, and a speech input module to be translated;
the method is characterized in that:
the to-be-translated voice input module is in remote wireless communication with a plurality of user terminals in different geographic positions and is used for receiving voice input signals sent by the user terminals;
the first number of first translation engines comprises a plurality of first translation engines for translating the voice to be translated into the voice of the target language; the second number of second translation engines comprises a plurality of second translation engines which translate the target language voice into language texts corresponding to the voice to be translated;
the second translation engine starts a plurality of threads when running, and each thread corresponds to one second translation engine;
the translation system determines whether to generate at least one abstract of the voice to be translated based on consensus feedback among the plurality of threads; the abstract comprises a user terminal number, a sending time and a voice input signal corresponding to the voice to be translated;
the to-be-translated voice input module comprises a voice preprocessing module, and the voice preprocessing module comprises an input voice grouping module; the input voice grouping module groups the voice to be translated based on the language type of the voice to be translated, and the language type of the language to be translated in each grouping is the same.
2. A multi-lingual translation system based on inter-thread consensus feedback as recited in claim 1, wherein:
the user terminal receives a voice input signal input by a user through a voice input interface, embeds geographic position information of the user terminal in the voice input signal, and sends the voice input signal to the voice input module to be translated as voice to be translated.
3. A multi-lingual translation system based on inter-thread consensus feedback according to claim 1 or 2 and wherein:
the voice preprocessing module further comprises a language identification module;
the language identification module identifies the language type of the current voice to be translated based on the geographic position information embedded in the voice to be translated.
4. A multi-lingual translation system based on inter-thread consensus feedback as recited in claim 3, wherein:
the first number is greater than the grouping number;
the input voice packet module comprises a plurality of packet data transmission channels, each packet data transmission channel being in unidirectional communication with a first translation engine;
the input voice grouping module sends the voice to be translated in each grouping to the corresponding first translation engine through a grouping data transmission channel.
5. A multi-lingual translation system based on inter-thread consensus feedback according to claim 1 or 2 or 4 and wherein:
the second number is less than or equal to the first number;
after the voice to be translated is input to the first translation engine, the first translation engine outputs a first translation result; and inputting the first translation result to a plurality of second translation engines, and outputting a plurality of second translation results by the plurality of second translation engines.
6. The multi-lingual translation system based on cross-thread consensus feedback as recited in claim 5, wherein:
the translation system determines whether to generate at least one abstract of the voice to be translated based on the common knowledge feedback among the threads, and specifically comprises the following steps:
performing similarity comparison on each second translation result and the voice to be translated to obtain a plurality of similarity comparison values;
and if each similarity comparison value is lower than a preset threshold value, generating at least one abstract of the voice to be translated.
7. The multi-lingual translation system based on cross-thread consensus feedback as recited in claim 6, wherein:
the translation system also comprises a voice text recognition system;
the voice text recognition system translates the voice to be translated into a text to be translated;
the second translation result comprises a plurality of second translation texts;
and comparing the similarity between the text to be translated and the plurality of second translation texts to obtain a plurality of similarity comparison values.
8. A method of multilingual translation based on inter-thread consensus feedback, the method being implemented based on the multilingual translation system based on inter-thread consensus feedback of any one of claims 1-7, the method comprising the steps of:
s810: starting a monitoring thread, wherein the monitoring thread judges whether voice access exists or not; if so, entering the next step;
s820: starting an identification thread, wherein the identification thread is used for identifying the language category of the voice accessed by the monitoring thread;
s830: starting a grouping thread, wherein the grouping thread is used for grouping the voice accessed by the monitoring thread;
s840: starting a plurality of first translation threads, wherein the plurality of first translation threads correspond to the language types of the voices accessed by the monitoring threads;
s850: starting a plurality of second translation threads, wherein the plurality of second translation threads are communicated with the plurality of first translation threads through a data pipeline, and obtaining translation results output by the first translation threads;
s860: starting a consensus thread, wherein the consensus thread performs consensus calculation based on the output results of the plurality of second translation threads and the voice accessed by the monitoring thread;
s870: judging whether a summary recording thread needs to be started or not based on the result of the consensus calculation; if yes, entering the next step; otherwise, returning to step S810;
s880: starting a summary recording thread, wherein the summary recording thread generates at least one summary of the voice accessed by the monitoring thread and records the at least one summary into a system file; the first translation thread comprises a plurality of first translation engines for translating the voice accessed by the monitoring thread into target language voice; the second translation thread comprises a plurality of second translation engines which can translate the target language voice into the voice corresponding language text accessed by the monitoring thread.
9. The method for multilingual translation based on cross-thread consensus feedback of claim 8 wherein: the abstract comprises a user terminal number, a sending time and a voice input signal which correspond to the voice accessed by the monitoring thread.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110034160.8A CN112818704B (en) | 2021-01-19 | 2021-01-19 | Multilingual translation system and method based on inter-thread consensus feedback |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110034160.8A CN112818704B (en) | 2021-01-19 | 2021-01-19 | Multilingual translation system and method based on inter-thread consensus feedback |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112818704A CN112818704A (en) | 2021-05-18 |
| CN112818704B true CN112818704B (en) | 2024-04-02 |
Family
ID=75869867
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110034160.8A Active CN112818704B (en) | 2021-01-19 | 2021-01-19 | Multilingual translation system and method based on inter-thread consensus feedback |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112818704B (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108595443A (en) * | 2018-03-30 | 2018-09-28 | 浙江吉利控股集团有限公司 | Simultaneous interpreting method, device, intelligent vehicle mounted terminal and storage medium |
| CN108628859A (en) * | 2018-04-25 | 2018-10-09 | 合肥上城信息技术有限公司 | A kind of real-time voice translation system |
| CN108664476A (en) * | 2018-05-08 | 2018-10-16 | 郑州大学西亚斯国际学院 | A kind of system and interpretation method of news translation Internet-based |
| CN109088995A (en) * | 2018-10-17 | 2018-12-25 | 永德利硅橡胶科技(深圳)有限公司 | Support the method and mobile phone of global languages translation |
| CN109446532A (en) * | 2018-09-11 | 2019-03-08 | 深圳市沃特沃德股份有限公司 | Translate bearing calibration, device and translation calibration equipment |
| CN109979461A (en) * | 2019-03-15 | 2019-07-05 | 科大讯飞股份有限公司 | A kind of voice translation method and device |
| CN110149805A (en) * | 2017-12-06 | 2019-08-20 | 创次源股份有限公司 | Double-directional speech translation system, double-directional speech interpretation method and program |
| CN110677406A (en) * | 2019-09-26 | 2020-01-10 | 上海译牛科技有限公司 | Simultaneous interpretation method and system based on network |
| KR20200095040A (en) * | 2019-01-31 | 2020-08-10 | 주식회사 네오픽시스 | Server and method for multilingual chat |
| CN111680524A (en) * | 2020-06-09 | 2020-09-18 | 语联网（武汉）信息技术有限公司 | Human-machine feedback translation method and system based on reverse matrix analysis |
| CN111753560A (en) * | 2020-06-28 | 2020-10-09 | 语联网（武汉）信息技术有限公司 | Self-adaptive translation system under parallel multi-source input mode |
| CN111950301A (en) * | 2020-08-25 | 2020-11-17 | 莱芜职业技术学院 | English translation quality analysis method and system for Chinese translation and English translation |
| CN112204653A (en) * | 2019-03-29 | 2021-01-08 | 谷歌有限责任公司 | Direct speech-to-speech translation through machine learning |
-
2021
- 2021-01-19 CN CN202110034160.8A patent/CN112818704B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110149805A (en) * | 2017-12-06 | 2019-08-20 | 创次源股份有限公司 | Double-directional speech translation system, double-directional speech interpretation method and program |
| CN108595443A (en) * | 2018-03-30 | 2018-09-28 | 浙江吉利控股集团有限公司 | Simultaneous interpreting method, device, intelligent vehicle mounted terminal and storage medium |
| CN108628859A (en) * | 2018-04-25 | 2018-10-09 | 合肥上城信息技术有限公司 | A kind of real-time voice translation system |
| CN108664476A (en) * | 2018-05-08 | 2018-10-16 | 郑州大学西亚斯国际学院 | A kind of system and interpretation method of news translation Internet-based |
| CN109446532A (en) * | 2018-09-11 | 2019-03-08 | 深圳市沃特沃德股份有限公司 | Translate bearing calibration, device and translation calibration equipment |
| CN109088995A (en) * | 2018-10-17 | 2018-12-25 | 永德利硅橡胶科技(深圳)有限公司 | Support the method and mobile phone of global languages translation |
| KR20200095040A (en) * | 2019-01-31 | 2020-08-10 | 주식회사 네오픽시스 | Server and method for multilingual chat |
| CN109979461A (en) * | 2019-03-15 | 2019-07-05 | 科大讯飞股份有限公司 | A kind of voice translation method and device |
| CN112204653A (en) * | 2019-03-29 | 2021-01-08 | 谷歌有限责任公司 | Direct speech-to-speech translation through machine learning |
| CN110677406A (en) * | 2019-09-26 | 2020-01-10 | 上海译牛科技有限公司 | Simultaneous interpretation method and system based on network |
| CN111680524A (en) * | 2020-06-09 | 2020-09-18 | 语联网（武汉）信息技术有限公司 | Human-machine feedback translation method and system based on reverse matrix analysis |
| CN111753560A (en) * | 2020-06-28 | 2020-10-09 | 语联网（武汉）信息技术有限公司 | Self-adaptive translation system under parallel multi-source input mode |
| CN111950301A (en) * | 2020-08-25 | 2020-11-17 | 莱芜职业技术学院 | English translation quality analysis method and system for Chinese translation and English translation |
Non-Patent Citations (2)
| Title |
|---|
| 基于多模态输入的多语种实时翻译软件设计与实现;权朝臣;邓长明;袁凌云;;云南师范大学学报(自然科学版);20200715(第04期);全文 * |
| 煤矿虚拟现实远程会议中实时翻译技术的应用研究;陈超;白磊;李芙玲;;华北科技学院学报(第06期);第51-55+64页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112818704A (en) | 2021-05-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110444196B (en) | Data processing method, device and system based on simultaneous interpretation and storage medium | |
| KR20170033722A (en) | Apparatus and method for processing user's locution, and dialog management apparatus | |
| US20140358516A1 (en) | Real-time, bi-directional translation | |
| CN110689877A (en) | Voice end point detection method and device | |
| US20220122622A1 (en) | Cascaded Encoders for Simplified Streaming and Non-Streaming ASR | |
| US20190121860A1 (en) | Conference And Call Center Speech To Text Machine Translation Engine | |
| CN110766442A (en) | Client information verification method, device, computer equipment and storage medium | |
| CN115485768A (en) | End-to-end multi-speaker overlapping speech recognition | |
| US20210312143A1 (en) | Real-time call translation system and method | |
| CN112818704B (en) | Multilingual translation system and method based on inter-thread consensus feedback | |
| CN107886940B (en) | Voice translation processing method and device | |
| CN112818709B (en) | Speech translation system and method for recording marks of multi-user speech conferences | |
| KR20230158107A (en) | Efficient streaming non-circular on-device end-to-end model | |
| CN116933806A (en) | Concurrent translation system and concurrent translation terminal | |
| KR20110065916A (en) | Interpretation system for error correction and auto scheduling | |
| US11881224B2 (en) | Multilingual speech recognition and translation method and related system for a conference which determines quantity of attendees according to their distances from their microphones | |
| CN112818702B (en) | Multi-user multi-language cooperative speech translation system and method | |
| CN113936655A (en) | Voice broadcast processing method and device, computer equipment and storage medium | |
| KR101233655B1 (en) | Apparatus and method of interpreting an international conference based speech recognition | |
| KR100399574B1 (en) | A system and method for interpreting automatically a telephony guidance for a foreigner | |
| CN112818705B (en) | Multilingual speech translation system and method based on group consensus | |
| CN112818708B (en) | System and method for processing voice translation of multi-terminal multi-language video conference in real time | |
| CN112818707B (en) | Reverse text consensus-based multi-turn engine collaborative speech translation system and method | |
| CN112818703B (en) | Multilingual consensus translation system and method based on multithread communication | |
| CN112818706B (en) | Voice translation real-time dispute recording system and method based on reverse result stability |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |