CN101957813A - Internet phone voice translation system and translation method - Google Patents
Internet phone voice translation system and translation method Download PDFInfo
- Publication number
- CN101957813A CN101957813A CN2009100233477A CN200910023347A CN101957813A CN 101957813 A CN101957813 A CN 101957813A CN 2009100233477 A CN2009100233477 A CN 2009100233477A CN 200910023347 A CN200910023347 A CN 200910023347A CN 101957813 A CN101957813 A CN 101957813A
- Authority
- CN
- China
- Prior art keywords
- translation
- language
- voice
- target
- unit
- 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.)
- Pending
Links
Images
Abstract
The invention relates to internet phone voice translation technology, in particular to a voice identification system and a voice translation method. The method has core ideas that the conventional internet phone is installed with software with voice translation function, and a translation service unit translates the information acquired through the voice identification system, sends the translation result to an opposite end and processes the voice information of the opposite end likewise so as to realize internet phone voice communication of different language users by the voice translation technology.
Description
Technical field
The present invention relates to the speech-sound intelligent translation technology, particularly a kind of networking telephone speech translation system and interpretation method.
Background technology
Because the now increasing people's custom of popularizing of network is communicated with each other by the mode of the networking telephone, but along with the trade of country country together, the continuous development and progress of interchange, the international trend of user must cause the country variant people to have the problem of aphasis when using the networking telephone to link up.With Chinese and English mother tongue user is example, can skillfully make in English the user few people of China to exchange with the American, and American also few people understands Chinese, so language becomes the biggest obstacle that internationalization exchanges.
The technology of also voice not being carried out instant translation in the existing networking telephone occurs, and networking telephone voice instant translation all will help people to carry out more convenient, accessible interchange widely.
Summary of the invention
The object of the present invention is to provide a kind of networking telephone speech translation system, the instant translation technology can be integrated on the diverse network phone, thus the communication between the iphone user of support different language.
Another object of the present invention is to provide a kind of multilingual instant translation is integrated into method on the networking telephone, can realize the translation conversion between the multilingual, thereby supports the communication between the iphone user of different language.
Be achieved in that in order to reach above purpose technical scheme of the present invention
(1) a kind of networking telephone voice instant translation system is characterized in that, comprising:
The multilingual translation Database Unit is used to store dictionary, short sentence storehouse, industry storehouse and the urtext information of different language and the translation contrast relationship between institute's Aim of Translation text message;
The translating server unit, the text translation after utilizing the multilingual translation Database Unit with speech conversion is target language text;
The terminal applies unit, the voice that to import or answer by the speech conversion unit are converted into text message, be translated as target language text through the translating server unit, after to be back to the speech conversion cell translation be that the target voice send via the network telephone signal transmitting terminal.
The characteristics of technique scheme are:
(a) described translating server unit comprises: analysis module, and synthesis module, sending module;
Wherein, analysis module: obtain and analyze the text message that becomes by speech conversion, from morphology, syntax data library unit, extract required urtext information and institute's Aim of Translation text message;
Wherein, synthesis module: will according to syntax rule and speech habits, synthesize target language from the text message and the Aim of Translation text message of morphology, the extraction of syntax data library unit;
Wherein, sending module: be used to send synthetic target language to Audio Processing Unit.
(b) described morphology, grammar database unit comprise: translation contrasting data storehouse is used to store urtext information that the speech conversion of input obtains and the translation contrast relationship between institute's Aim of Translation text message; Described translation contrasting data storehouse comprise first language to second language contrasting data storehouse and second language to first language contrasting data storehouse.
(c) described terminal applies unit further selectivity comprises:
The target interpretive language setting of input voice and the beginning and the end mark of voice paragraph.
Typing or answer the module that is provided with of voice can be provided with the word speed speed and the volume of voice.
(2) a kind of networking telephone voice instant translation method, it is characterized in that, may further comprise the steps: after the terminal user configures target language, is text message by Audio Processing Unit with terminal user's speech conversion, be target language text by the translation service unit with the text translation that obtains again, and return Audio Processing Unit and be treated to the target voice, Audio Processing Unit sends to the Zhongdao opposite end of network telephone signal transmit port networking telephone receiver with the target sound result; Or the terminal user hears the opposite end voice messaging, and at first the opposite end voice are carried out text-converted is target text by the translation service cell translation to Audio Processing Unit then, handles outputing to terminal user's receiver at last by Audio Processing Unit.
Description of drawings
Fig. 1 is a kind of networking telephone voice instant translation system figure
Fig. 2 is a networking telephone voice instant translation data flow diagram
Embodiment
Core concept of the present invention is: the voice translation functions in the existing network phone between integrated different language, by the translating server unit voice messaging of input user typing is translated, or the voice messaging that end subscriber is sent translated, thereby realize the support that the iphone user to different language converses.
With reference to Fig. 1, networking telephone speech translation system of the present invention comprises: translating server unit, terminal applies unit.The translating server unit further comprises: CPU (central processing unit) 4, analysis and processing unit 5, synthesis unit 7, morphology, syntax data library unit 6.Wherein, unit 4 is used to handle the translation request of self terminal, and returns translation result after translation is finished.Unit 5 obtains and analyzing speech conversion urtext, according to analysis result, according to processing rule, extracts required text message from morphology, grammar database.The text message that unit 7 will extract from database, according to syntax rule, and speech habits, synthetic target text.Unit 6, morphology, grammar database comprise, translation contrasting data storehouse comprise first language to second language contrasting data storehouse and, second language contrasts the storehouse to first language.The terminal applies unit further comprises: target language selected cell 1, voice-input unit 2, Audio Processing Unit 3.Wherein, unit 1 is used to be provided with the target interpretive language; Unit 2 is used to judge the beginning and the end of phonetic entry paragraph; Unit 3 is used for the voice of input end are transformed into text message, simultaneously the Aim of Translation language text is transformed into voice and sends to the network telephone signal transmitting terminal.
Among the present invention, the translation contrast relationship of described database 6 storages can be stored in the translation table of comparisons, the described translation table of comparisons comprises: first language is to the translation table of comparisons of second language or the second language translation table of comparisons to first language, also can have first language simultaneously to the translation table of comparisons of second language and the second language translation table of comparisons to first language.Accordingly, the translation unit 6 of the networking telephone can only have the interpretative function of first language to the interpretative function of second language or second language to first language, also can have the interpretative function of first language to the interpretative function of second language and second language to first language simultaneously.
Here, described first language can be a Chinese, English or the like; Corresponding with it, second language can be English, Chinese or the like.So, the translation table of comparisons of storing in the described database is: the translation table of comparisons between Chinese and the foreign language is used for translator of Chinese is become foreign language; Or be the translation table of comparisons between foreign language and the Chinese, be used for foreign languages translation is become Chinese; Or, translator of Chinese can be become foreign language for the translation table of comparisons of asking intertranslation of Chinese with foreign language, foreign languages translation can be become Chinese again.
Among the present invention, database 6 main storages are based on the class libraries of language material, and so-called class libraries based on language material is meant: contain the urtext information of various sentence patterns and the language material class libraries of target text information bilingual journal based on being provided with one.When translation, the similar sentence of sentence in extraction and the input original speech information imitates example sentence then and realizes that original speech information arrives the conversion of target voice messaging from the language material class libraries.And because the corresponding dictionary of storage in the database, therefore, the kind of interpretive language can be selected according to the heterogeneous networks telephone subscriber.Here, the heterogeneous networks telephone subscriber can be the user of country variant.If what original speech information adopted is first language, then the target voice messaging adopts second language; Otherwise if the original speech information employing is second language, then the target voice messaging adopts first language.
For any two iphone users in the networking telephone real-time phonetic translation system, what send voice messaging can be described as local terminal real-time phonetic terminal, receiving speech information can be described as opposite end real-time phonetic terminal, local terminal real-time phonetic terminal links to each other by Internet with opposite end real-time phonetic terminal.In in transcription platform of the present invention, can be in a networking telephone integrated interpretative function, also can be at two all integrated interpretative functions of the networking telephone.That is to say that terminal applies recited above unit can only be arranged in the local terminal real-time phonetic terminal or only be arranged in the opposite end speech communication terminal, also can be arranged at simultaneously in local terminal real-time phonetic terminal and the opposite end real-time phonetic terminal.
In the transcription platform, any one real-time phonetic terminal applies can adopt structure shown in Figure 1 among the present invention.Specifically, terminal applies of the present invention can adopt following several basic framework in actual applications: 1. local terminal real-time phonetic terminal adopts structure shown in Figure 1, and the opposite end networking telephone remains unchanged; 2. local terminal real-time phonetic terminal remains unchanged, and opposite end real-time phonetic terminal adopts structure shown in Figure 1; 3. local terminal real-time phonetic terminal and opposite end real-time phonetic terminal all adopt structure shown in Figure 1, that is: each instant communication terminal can both be supported the translation of first language to the translation of second language and second language to first language.
For each real-time phonetic terminal that adopts Fig. 1 structure, can also further be provided with select to send module is set, translation is provided with module etc.
Based on above-mentioned multilingual translation platform, multilingual translation platform method provided by the present invention is: earlier original speech information is translated as the target voice messaging, sends.
Two kinds of implementations are specifically arranged: a kind of mode is, local terminal real-time phonetic terminal receives the original speech information of user's input, and original speech information is translated as the target voice messaging, afterwards, the target voice messaging is sent to opposite end real-time phonetic terminal by the networking telephone.Another kind of mode is: local terminal real-time phonetic terminal receives the original speech information that opposite end real-time phonetic terminal is sent, and original speech information is translated as the target voice messaging, afterwards, the target voice messaging is offered this end subscriber.
At two kinds of implementations of above-mentioned multilingual real-time phonetic method, this method further comprises: set in advance the target language that need translate;
Specifically, be provided with to determine the language form of transmitting terminal and opposite end by the function of network phone key before voice send, the voice conversion unit can be changed automatically according to initial setting when sending voice and receiving speech information like this.
In the implementation of above-mentioned networking telephone real-time phonetic interpretation method, described translation is specially: local terminal analysis and processing unit 5, to handle the text message that imports into through CPU (central processing unit) by Audio Processing Unit, then the result is sent into morphology, syntax data library unit by analyzing after by the synthetic target language text information of synthesis unit.The result of synthesis unit enters the synthetic at last target voice of Audio Processing Unit by CPU (central processing unit), sends to end subscriber by the networking telephone at last.
Here, analysis and processing unit 5 is obtained the text message of raw tone correspondence, and according to the urtext information of storage in the database 6 and the translation contrast relationship between the target text information, urtext information is translated as the process of target text information, adoptable interpretative system has multiple, such as: both can adopt the interpretative system that belongs to traditional, comprise: direct translation method (Direct Translation), interlingual approach (Intel Lingual Approach), transformation approach (Transfer Approach); Can adopt based on class libraries again, belong to the interpretative system of modernist
Wherein, directly translation method is meant: the word in the text message of raw tone correspondence, fixed phrases or sentence direct replacement are become tie element in the target text information.
Interlingual approach is meant: earlier the word in the text message of raw tone correspondence, fixed phrases or sentence are analyzed, convert a kind of sentence structure that all target text information all is fit to again to, that is: semantic expressiveness, and generate any target text information according to semantic expressiveness.
Transformation approach is meant: adopt two kinds of internal representations and translate by three phases, first stage converts the text message of raw tone correspondence to the internal representations of the text message of raw tone correspondence, subordinate phase converts the internal representations of the text message of raw tone correspondence the internal representations of target text information to, and the phase III generates target text information according to the internal representations of target text information again.
In the networking telephone voice translation method of the present invention, when existing the plurality of target language available, before inquiring about, also need first setting to translate to set out language and target language, afterwards, analysis and processing unit is inquired about according to being set in the corresponding database.Such as: Set For Current has English database, Korean database, if the target language of Set For Current is English, then analysis and processing unit can be that rope is inquired about in English database with the text message of raw tone correspondence, obtains target text information.Usually, can determine the target language of a kind of language,,, all be translated as English when then translating if target language is not set such as English for acquiescence.
Embodiment one:
In the present embodiment, first language is a Chinese, and original speech information adopts the Chinese speech input; Second language is English, and the target voice messaging adopts English equivalents.Described interpretative function is finished by the translation of local terminal real-time phonetic translating server before sending voice, and described translation unit and database are arranged in the translating server, and the setting of Aim of Translation language is defaulted as English, that is: be English with translator of Chinese.That is to say that transmit leg provides interpretative function in the present embodiment, the take over party does not provide interpretative function.
Figure 2 shows that the networking telephone voiced translation platform that present embodiment adopts, the local terminal that sends voice adopts real-time phonetic translation structure shown in Figure 1.So, when the original speech information of local terminal real-time phonetic typing was " good morning ", the treatment scheme of voiced translation platform method was in the present embodiment:
Step 1, local user be by networking telephone typing voice messaging " good morning ", and the Audio Processing Unit 1 that this moment, voice messaging entered into Fig. 2 converts corresponding text message to.
Text message after step 2, Audio Processing Unit 1 will be changed is submitted to CPU (central processing unit) 2.
At this moment, reply after hearing voice messaging the opposite end, when the Audio Processing Unit of opposite end local terminal in Recovery Process is that basis for estimation comes labeled statement to finish to give simultaneously prompt tone of end subscriber is represented please wait in the translation with the dead time length in the statement of opposite end.Behind step 1,2,3 translation process, the answer of opposite end will be with the linguistic form output voice of local terminal.Thereby the timely voice of finishing between the different language user are linked up.
The above only is preferred embodiment of the present invention the sixth of the twelve Earthly Branches, is not to be used to limit protection scope of the present invention.
Below be the subprogram code that voice are converted into text message, the program development language is JAVA.
Public?class?MyRec{
// create recognizer and resource is distributed
Recognizer?recognizer=Central.createRecognizer(null);recognizer.allocate();
// interpolation engine audiomonitor-represent that when the typing voice begin, and stop or the like
recognizer.addEngineListener(engineListener);
// read grammar file
File?gf=new?FileReader(grammarFile);
RuleGrammar?rules=recognizer.loadJSGF(gf);
// interpolation is audiomonitor as a result
rules.addResultListener(new?VoiceListener());
// speech recognizer processes changes
recognizer.commitChanges();
recognizer.requestFocus();
// begin to monitor
recognizer.resume();
……
}
Below be that acceptance and processed voice are the code snippet of text
public?class?VoiceListener?extends?ResultAdaptor{
public?void?resultAccepted(ResultEvent?re){
FinalRuleResult?result=(FinalRuleResult)re.getSource();
// obtain one to check grammatical markers
String[]tags=result.getTags();
System.out.println(″First?tag?was:″+tags[0]);
}
……
}
Claims (6)
1. a networking telephone voiced translation technology is characterized in that, comprising:
Audio Processing Unit, the speech conversion that is used for input is a text message, does information input preparation for entering the translation service unit; The target text that will obtain after will changing simultaneously converts corresponding target voice to;
Morphology, syntax data library unit are used to store dictionary, short sentence storehouse, industry storehouse and the urtext information of different language and the translation contrast relationship between institute's Aim of Translation text message;
The translating server unit utilizes multilingual morphology, syntax data library unit that the text message that the speech conversion unit obtains is translated as target language.
2. a kind of networking telephone speech translation system according to claim 1 is characterized in that described translating server unit comprises: analysis module, and synthesis module, sending module;
Wherein, analysis module: obtain and analyze the text message that becomes by speech conversion, from morphology, syntax data library unit, extract required urtext information and institute's Aim of Translation text message;
Wherein, synthesis module: will according to syntax rule and speech habits, synthesize target language from the text message and the Aim of Translation text message of morphology, the extraction of syntax data library unit;
Wherein, sending module: be used to send synthetic target language to Audio Processing Unit.
3. a kind of networking telephone speech translation system according to claim 1, it is characterized in that, described morphology, grammar database unit comprise: translation contrasting data storehouse is used to store urtext information that the speech conversion of input obtains and the translation contrast relationship between institute's Aim of Translation text message; Described translation contrasting data storehouse comprise first language to second language contrasting data storehouse and second language to first language contrasting data storehouse.
4. a kind of networking telephone speech translation system according to claim 1 is characterized in that, sends voice module is set, and the word speed speed and the volume of voice is set.
5. networking telephone voice translation method, it is characterized in that, may further comprise the steps: after the terminal user configures target language, is text message by Audio Processing Unit with terminal user's speech conversion, be target language text by the translation service unit with the text translation that obtains again, and return Audio Processing Unit and be treated to the target voice, Audio Processing Unit sends to the Zhongdao opposite end of transmit port with the target sound result; Or the terminal user hears the opposite end voice messaging, and at first the opposite end voice are carried out text-converted is target text by the translation service cell translation to Audio Processing Unit then, handles outputing to the terminal user at last by Audio Processing Unit.
6. a kind of networking telephone voice translation method according to claim 5 is characterized in that, before the input target voice of the voice messaging of typing, target language, conversion is set earlier, sends again.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2009100233477A CN101957813A (en) | 2009-07-16 | 2009-07-16 | Internet phone voice translation system and translation method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2009100233477A CN101957813A (en) | 2009-07-16 | 2009-07-16 | Internet phone voice translation system and translation method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101957813A true CN101957813A (en) | 2011-01-26 |
Family
ID=43485146
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2009100233477A Pending CN101957813A (en) | 2009-07-16 | 2009-07-16 | Internet phone voice translation system and translation method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101957813A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102902666A (en) * | 2011-07-26 | 2013-01-30 | 郑俨 | Multi-country speech recognition and translation screen display system applied to network protocol speech |
CN103810158A (en) * | 2012-11-07 | 2014-05-21 | 中国移动通信集团公司 | Speech-to-speech translation method and device |
CN104679733A (en) * | 2013-11-26 | 2015-06-03 | 中国移动通信集团公司 | Voice conversation translation method, device and system |
CN105511601A (en) * | 2014-10-08 | 2016-04-20 | Lg电子株式会社 | Mobile terminal and controlling method thereof |
CN107015970A (en) * | 2017-01-17 | 2017-08-04 | 881飞号通讯有限公司 | A kind of method that bilingual intertranslation is realized in network voice communication |
CN107066453A (en) * | 2017-01-17 | 2017-08-18 | 881飞号通讯有限公司 | A kind of method that multilingual intertranslation is realized in network voice communication |
CN108305630A (en) * | 2018-02-01 | 2018-07-20 | 中科边缘智慧信息科技(苏州)有限公司 | Language transmission method under low-bandwidth condition and speech transmission index |
CN109977429A (en) * | 2019-04-03 | 2019-07-05 | 新疆语视未来信息科技有限公司 | A kind of information interacting method based on translation content instant playback |
-
2009
- 2009-07-16 CN CN2009100233477A patent/CN101957813A/en active Pending
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102902666A (en) * | 2011-07-26 | 2013-01-30 | 郑俨 | Multi-country speech recognition and translation screen display system applied to network protocol speech |
CN103810158A (en) * | 2012-11-07 | 2014-05-21 | 中国移动通信集团公司 | Speech-to-speech translation method and device |
CN104679733A (en) * | 2013-11-26 | 2015-06-03 | 中国移动通信集团公司 | Voice conversation translation method, device and system |
CN105511601A (en) * | 2014-10-08 | 2016-04-20 | Lg电子株式会社 | Mobile terminal and controlling method thereof |
CN105511601B (en) * | 2014-10-08 | 2020-05-05 | Lg电子株式会社 | Mobile terminal and control method thereof |
CN107015970A (en) * | 2017-01-17 | 2017-08-04 | 881飞号通讯有限公司 | A kind of method that bilingual intertranslation is realized in network voice communication |
CN107066453A (en) * | 2017-01-17 | 2017-08-18 | 881飞号通讯有限公司 | A kind of method that multilingual intertranslation is realized in network voice communication |
CN108305630A (en) * | 2018-02-01 | 2018-07-20 | 中科边缘智慧信息科技(苏州)有限公司 | Language transmission method under low-bandwidth condition and speech transmission index |
CN109977429A (en) * | 2019-04-03 | 2019-07-05 | 新疆语视未来信息科技有限公司 | A kind of information interacting method based on translation content instant playback |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101867632A (en) | Mobile phone speech instant translation system and method | |
CN101957814A (en) | Instant speech translation system and method | |
CN101957813A (en) | Internet phone voice translation system and translation method | |
CN110111780B (en) | Data processing method and server | |
CN1333385C (en) | Voice browser dialog enabler for a communication system | |
CN101207586B (en) | Method and system for real-time automatic communication | |
CN101072168B (en) | Multi-language instant communication terminal and its system and method | |
CN101923858B (en) | Real-time and synchronous mutual translation voice terminal | |
US7593842B2 (en) | Device and method for translating language | |
US9058322B2 (en) | Apparatus and method for providing two-way automatic interpretation and translation service | |
CN101494621A (en) | Translation system and translation method for multi-language instant communication terminal | |
CN1323435A (en) | System and method for providing network coordinated conversational services | |
CN101291336A (en) | System and method for concurrent multimodal communication | |
CN201298231Y (en) | Multilingual communication and application system capable of automatically identifying multilanguage | |
CN103533129B (en) | Real-time voiced translation communication means, system and the communication apparatus being applicable | |
CN110517668B (en) | Chinese and English mixed speech recognition system and method | |
CN104125548A (en) | Method of translating conversation language, device and system | |
JP2017120616A (en) | Machine translation method and machine translation system | |
CN101834809A (en) | Internet instant message communication system | |
CN108810187B (en) | Network system for butting voice service through block chain | |
CN114064943A (en) | Conference management method, conference management device, storage medium and electronic equipment | |
CN108447473A (en) | Voice translation method and device | |
CN115455981B (en) | Semantic understanding method, device and equipment for multilingual sentences and storage medium | |
CN116189663A (en) | Training method and device of prosody prediction model, and man-machine interaction method and device | |
CN112818709B (en) | Speech translation system and method for recording marks of multi-user speech conferences |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20110126 |