CN108628851A - The method for translating mandarin and Japanese based on artificial intelligence algorithm of support vector machine - Google Patents
The method for translating mandarin and Japanese based on artificial intelligence algorithm of support vector machine Download PDFInfo
- Publication number
- CN108628851A CN108628851A CN201710174495.3A CN201710174495A CN108628851A CN 108628851 A CN108628851 A CN 108628851A CN 201710174495 A CN201710174495 A CN 201710174495A CN 108628851 A CN108628851 A CN 108628851A
- Authority
- CN
- China
- Prior art keywords
- japanese
- vector machine
- support vector
- mandarin
- translation
- 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
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
- G06F40/42—Data-driven translation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/16—Sound input; Sound output
- G06F3/162—Interface to dedicated audio devices, e.g. audio drivers, interface to CODECs
-
- 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 discloses a kind of methods for translating mandarin and Japanese based on artificial intelligence algorithm of support vector machine, including:1)Audio input device;2)Audio output apparatus;3)The mandarin audio large database concept of acquisition;4)The Japanese audio large database concept of acquisition;5)By Hiragana and katakana permutation and combination and its large database concept of paraphrase and syntactic rule;6)The large database concept of the Chinese written language structure and word constituent grammars that be made of radical radical;7)The translation model based on algorithm of support vector machine built in large data center, pass through above-mentioned component, the present invention can substitute the advanced Sino-Japan simultaneous interpretation translation of high wages, provide that cheap and can not fear that fatigue can carry out long-time high quality translation can translate into mandarin Japanese or the translation service by Japanese Translator at mandarin to the user.
Description
Technical field
The present invention relates to the fields that algorithm of support vector machine is used for speech recognition and translation, more particularly to based on artificial intelligence
The method of energy algorithm of support vector machine translation mandarin and Japanese.
Background technology
With the quickening of internationalization process, the demand of translation is increasing, and existing simultaneous interpretation translation is completed by people,
Simultaneous interpretation translator's labor intensity of profession is big, and translation accuracy is vulnerable to the influence of personal physical factors, in international conference,
If the duration of meeting is long, after the physical and energy constantly overdraw of translator, it will because fatigue makes the accurate of translation
Degree declines;When individual travels abroad, since the simultaneous interpretation translation salary level of profession is high, general ruck is relatively difficult to receive to take
Band translator goes on a journey.
Invention content
It is common based on the translation of artificial intelligence algorithm of support vector machine that the invention mainly solves the technical problem of providing one kind
The method of words and Japanese, can substitute the high level translation of high wages, and providing to the user will not be because of translation time length and because of fatigue
Caused translation error.
In order to solve the above technical problems, one aspect of the present invention is:It provides a kind of based on artificial intelligence branch
Hold vector machine algorithm translation mandarin and the method for Japanese, which is characterized in that including:1)The audio input device of mandarin, 2)
Translate into Japanese audio output apparatus, 3)The mandarin audio large database concept of acquisition, 4)The Japanese audio big data of acquisition, 5)
By Hiragana and katakana permutation and combination and its large database concept of paraphrase and syntactic rule, 6)The Chinese being made of radical radical
Language text structure and the large database concept of word constituent grammar, 7)The turning over based on algorithm of support vector machine built in large data center
Translate model;By above-mentioned seven components, the present invention can substitute the advanced Sino-Japan simultaneous interpretation translation of high wages, provide price to the user
It is cheap and can not fear fatigue can carry out long-time high quality translation mandarin can be translated into Japanese or by Japanese Translator at
The translation of mandarin.
Based on the method that artificial intelligence algorithm of support vector machine translates mandarin and Japanese, concrete mode is:
Supporting vector gives linearly inseparable training dataset
Wherein, ,, the linear SVM study of linearly inseparable can be of equal value
Ground solves corresponding convex quadratic programming problem, and form is such as:
Acquire optimal solutionWith
Hyperplane to be detached:
Categorised decision function
Wherein,For punishment parameter, after the calculated best translated speech of algorithm of support vector machine, through internet by audio
Data are transferred to audio output device, play to user in real time and listen, and translation is completed.
Specific implementation mode
In one embodiment, the user A to speak standard Chinese pronunciation says a mandarin against translater audio input device, leads to
Network is crossed by the algorithm of support vector machine model of the transmission of speech information to cloud computing center, with the big data after deep learning
After being compared, the audio-frequency information synchronous transfer of Japanese will be translated into translater audio output apparatus, user B uses the equipment
The Japanese pronunciation of the translation of the simultaneous interpretation to user's A speech contents is heard.
In another embodiment, it says that the user B of Japanese says a Japanese against translater audio input device, passes through
Network by the algorithm of support vector machine model of the transmission of speech information to cloud computing center, with the big data after deep learning into
After row compares, the audio-frequency information synchronous transfer of mandarin will be translated into translater audio output apparatus, user A uses the equipment
The translation audio of the mandarin of the translation of the simultaneous interpretation to user's B speech contents is heard.
Claims (4)
1. the method for translating mandarin and Japanese based on artificial intelligence algorithm of support vector machine, which is characterized in that including:1)Audio
Input equipment;2)Audio output apparatus;3)The mandarin audio large database concept of acquisition;4)The Japanese audio large database concept of acquisition;
5)By Hiragana and katakana permutation and combination and its large database concept of paraphrase and syntactic rule;6)It is made of radical radical
The large database concept of Chinese written language structure and word constituent grammar;7)Built in large data center based on algorithm of support vector machine
Translation model, seven components.
2. the method according to claim 1 that mandarin and Japanese are translated based on artificial intelligence algorithm of support vector machine,
It is characterized in that:Component is divided into user terminal physical components and server-side cloud computing component is constituted;User's end pieces are claim 1 institute
1 stated)With 2);Server-side cloud computing component is described in claim 13)、4)、5)、6)、7), and component 7)It needs to component
3)、4)、5)、6)It could be to component 1 after the deep learning of the algorithm of support vector machine of progress big data)The voice data that input comes
It is translated, then passes through component 2)By the voice transfer after translation to component 2).
3. the method according to claim 1 that mandarin and Japanese are translated based on artificial intelligence algorithm of support vector machine,
It is characterized in that including such as step:
Step 1: Japanese character and grammer big data are acquired with Chinese written language and grammer big data;
Step 2: japanese voice big data is acquired with Chinese speech big data;
Step 3: all data are scanned in large database concept, and after taxonomic revision, the support vector machines of typing cloud computing center is calculated
Method model;
Step 4: carrying out deep learning by algorithm of support vector machine centering day translation data:Component 1 described in claim 1)
Middle input makes it through after support vector machines translation model is translated not less than 10000 Japanese audios from described in claim 1
Component 2)Middle output audio detects it and translates accuracy;It is described in claim 1 that 10000 mandarins inputs will be not less than again
Component 1)From component 2 described in claim 1 after being translated by support vector machines translation model)Middle output audio, detects it and turns over
Translate accuracy;If the above-mentioned friendship detected twice, which passes translation accuracy rate, is higher than 95%, the accuracy rate of simultaneous interpretation translation is supported higher than 70%
Vector machine model is trained successfully, can be come into operation;If accuracy rate is relatively low, repeatedly step 3 is to step 6, and extends support
The deep learning time of vector machine model, until terminating after translation accuracy rate is up to standard.
4. the method according to claim 1 for translating mandarin and Japanese based on artificial intelligence algorithm of support vector machine, branch
Holding specific method of the vector machine translation model for calculating translated speech is:
Supporting vector gives linearly inseparable training dataset
Wherein, ,, the linear SVM study of linearly inseparable can be of equal value
Ground solves corresponding convex quadratic programming problem, and form is such as:
Acquire optimal solutionWith
Hyperplane to be detached:
Categorised decision function
Wherein,For punishment parameter.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710174495.3A CN108628851A (en) | 2017-03-22 | 2017-03-22 | The method for translating mandarin and Japanese based on artificial intelligence algorithm of support vector machine |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710174495.3A CN108628851A (en) | 2017-03-22 | 2017-03-22 | The method for translating mandarin and Japanese based on artificial intelligence algorithm of support vector machine |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108628851A true CN108628851A (en) | 2018-10-09 |
Family
ID=63707086
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710174495.3A Pending CN108628851A (en) | 2017-03-22 | 2017-03-22 | The method for translating mandarin and Japanese based on artificial intelligence algorithm of support vector machine |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108628851A (en) |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102811284A (en) * | 2012-06-26 | 2012-12-05 | 深圳市金立通信设备有限公司 | Method for automatically translating voice input into target language |
-
2017
- 2017-03-22 CN CN201710174495.3A patent/CN108628851A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102811284A (en) * | 2012-06-26 | 2012-12-05 | 深圳市金立通信设备有限公司 | Method for automatically translating voice input into target language |
Non-Patent Citations (1)
Title |
---|
张春祥 等: "《基于短语评价的翻译知识获取》", 29 February 2012, 哈尔滨:哈尔滨工业大学出版社 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Caubrière et al. | Curriculum-based transfer learning for an effective end-to-end spoken language understanding and domain portability | |
CN108647214A (en) | Coding/decoding method based on deep-neural-network translation model | |
US9760565B2 (en) | Natural expression processing method, processing and response method, device, and system | |
CN111477216B (en) | Training method and system for voice and meaning understanding model of conversation robot | |
CN107066455A (en) | A kind of multilingual intelligence pretreatment real-time statistics machine translation system | |
CN109977398B (en) | Speech recognition text error correction method in specific field | |
CN108287820A (en) | A kind of generation method and device of text representation | |
CN106803422A (en) | A kind of language model re-evaluation method based on memory network in short-term long | |
CN1731510B (en) | Text-speech conversion for amalgamated language | |
Kumar et al. | Translations of the CALLHOME Egyptian Arabic corpus for conversational speech translation | |
CN112463942A (en) | Text processing method and device, electronic equipment and computer readable storage medium | |
CN104679733B (en) | A kind of voice dialogue interpretation method, apparatus and system | |
Min et al. | Exploring the integration of large language models into automatic speech recognition systems: An empirical study | |
CN104217039A (en) | Method and system for recording telephone conversations in real time and converting telephone conversations into declarative sentences | |
CN110852075B (en) | Voice transcription method and device capable of automatically adding punctuation marks and readable storage medium | |
CN109859746B (en) | TTS-based voice recognition corpus generation method and system | |
CN108628851A (en) | The method for translating mandarin and Japanese based on artificial intelligence algorithm of support vector machine | |
CN103268314B (en) | A kind of method and device obtaining Thai language punctuate rule | |
EP4152280A3 (en) | Method and apparatus for recognizing text, and method and apparatus for training text recognition model | |
CN108628841A (en) | The APP of Guangdong language accent and English is translated based on BIRCH clustering algorithms | |
CN108717854A (en) | Method for distinguishing speek person based on optimization GFCC characteristic parameters | |
CN108628847A (en) | A kind of simultaneous interpretation case for translating mandarin and English using BIRCH clustering algorithms | |
Rayner et al. | A framework for rapid development of limited-domain speech-to-sign phrasal translators | |
CN108628848A (en) | The method that Sichuan accent and English are translated with BIRCH clustering algorithms | |
CN104966513B (en) | Verbal order treating method and apparatus |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20181009 |
|
WD01 | Invention patent application deemed withdrawn after publication |