CN101488342A - Human-machine language interaction deduction system and intelligent implementing method for human-machine language interaction demand response - Google Patents
Human-machine language interaction deduction system and intelligent implementing method for human-machine language interaction demand response Download PDFInfo
- Publication number
- CN101488342A CN101488342A CNA2008102206499A CN200810220649A CN101488342A CN 101488342 A CN101488342 A CN 101488342A CN A2008102206499 A CNA2008102206499 A CN A2008102206499A CN 200810220649 A CN200810220649 A CN 200810220649A CN 101488342 A CN101488342 A CN 101488342A
- Authority
- CN
- China
- Prior art keywords
- language
- module
- voice
- result
- data
- 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 discloses a man-machine language interactive deduction system and a man-machine language interactive requirement response intellectualization implementation method. The system comprises a speech recognition processing module, an interactive voice response module including a text-to-speech conversion module, a service processing module, a service control module and a database composed of a language vocabulary sentence pattern base, a language historical material base, a language information base and a service database. The system and method of the invention have the advantages of solving the deficiency and difficulty of the existing speech recognition technology, solving the homophone recognition and error correction difficulty in the speech recognition technology application and text-to-speech conversion technology, solving multichannel automatic flow control technology difficulty in collecting speech requirements during man-computer speech interaction, solving the technology difficulty that machine can understand languages and analyze response results and solving the technology difficulty in matching processing of results of the man-machine language interactive requirement response.
Description
Technical field
The present invention relates to speech recognition technology and the intelligentized technology of man machine language's interaction demand response.
Background technology
Speech recognition technology and man machine language's interaction technique just existed and use in last century.But have a lot of shortcomings and technological difficulties from operational angle, and can not the degree of depth and used by market widely.Mainly there is following problem in it:
(1), speech recognition technology is applied in ' phonitic entry method ', what allow people feel is that the speech recognition error rate is too high, can't correctly judge affirmation to phonetically similar word, malapropism wrongly written character, causes this service application stagnation thus;
(2), when man machine language's interaction technique is applied in the voice call origination function of the communications field, its main means only are to realize that by terminal device ' voice ' finish target processing to the variation of ' literal order '.Ultimate principle and realization also are like this when being applied in the information service aspect.
(3), the technological difficulties of man machine language's interaction technique maximum are that people's demand is drawn and can't be understood and analyze, the language that the people is said can't remove interpretive semantic and pragmatic.Do for example the people asks during man-machine interaction: you understand guessing riddles? can you carry on the back Tang poetry? or other intellectualities the problem of meeting thinking metalanguage, interpretive semantic the time just seem powerless.
Summary of the invention
The objective of the invention is existence, a kind of intelligentized meeting thinking metalanguage and the mutual deduction system of man-machine language of interpretive semantic and the intelligent implementation method of man-machine language interaction demand response are provided at the problems referred to above.
The objective of the invention is to be achieved through the following technical solutions:
The mutual deduction system of a kind of man-machine language, comprise voice acquisition module ASR sound identification module, Service Processing Module, the interactive voice response module that includes literary composition language modular converter, message control module and database, wherein, voice acquisition module is to gather multichannel people's language tone signal and it is delivered to ASR voice recognition processing module, described voice recognition processing module is utilized computer technology and logic of language, acoustic feature principle and phonetic feature principle are as instrument processed voice identification process, repeatedly unclog and readjust and obtain voice identification result and this voice identification result is delivered to Service Processing Module by language vocabulary being intersected interaction, Service Processing Module utilizes logic of language as instrument, according to mathematics, physics, computer science, philosophy, the science of law, linguistics etc. are as application foundation, at language shape, semantic, pragmatic makes an explanation and distributes and submit to the result to message control module, message control module will carry out the processing of professional control flow from the data result that Service Processing Module obtains, and described message control module comprises traffic assignments, business audit, professional output module, wherein, described service distribution module is according to the data result that obtains in the Service Processing Module, distributes to be submitted to language historical data storehouse and to mate the data result that is obtained with Service Database and carry out business audit; Described professional output module is the literary composition language modular converter of giving professional output module with the data delivery of business audit and the result being exported to the interactive voice response module, the interactive voice response module is carried out hyperchannel automatic flow control and treatment to the voice demand collection in the interactive voice flow process, and utilize its literary composition language modular converter to enter prosody modeling, text analyzing, phonetic synthesis, literary composition language correction process and realize interactive voice demand response result's flow processing is finished in the processing that text receives, voice shift, result's calibration, result send voice.
Wherein above-mentioned database comprises the phonetic feature storehouse of being made up of voice messaging storehouse and language vocabulary sentence pattern storehouse, language historical data storehouse, Service Database, wherein said phonetic feature storehouse is to utilize phonetic feature modeling image data to finish, described voice messaging storehouse is to utilize the man-machine interaction language message to be stored into the language message storehouse after handling audit, described language vocabulary sentence pattern storehouse is to utilize language vocabulary, the dialect of the multiple popular constant of sentence pattern and collection and the data of phonetic synthesis compilation thereof are finished, described language historical data storehouse comprises: the mathematics historical data, the philosophy historical data, the physics historical data, the chemistry historical data, multidisciplinary historical data such as language historical data, reference book compilation is finished, and described Service Database comprises that the data compilation of business demand collection finishes.
The mutual deduction system of man-machine language carries out the intelligent implementation method of man-machine language interaction demand response, may further comprise the steps:
(1), voice acquisition module is delivered to ASR voice recognition processing module with the hyperchannel people language tone signal that collects;
(2), ASR voice recognition processing module utilizes computer technology and logic of language, acoustic feature principle and phonetic feature principle as instrument processed voice identification process, repeatedly unclog and readjust and obtain voice identification result and this voice identification result is delivered to Service Processing Module by language vocabulary being intersected interaction;
(3), Service Processing Module utilizes logic of language as instrument, as application foundation, at distributions that make an explanation of language shape, semanteme, pragmatic, the result is to message control module in submission according to mathematics, physics, computer science, philosophy, the science of law, linguistics etc.;
(4), message control module will carry out the processing of professional control flow from the data result that Service Processing Module obtains, and described message control module comprises traffic assignments, business audit, professional output module, wherein, described service distribution module is according to the data result that obtains in the Service Processing Module, distributes to be submitted to language historical data storehouse and to mate the data result that is obtained with Service Database and carry out business audit; Described professional output module is the literary composition language modular converter of giving professional output module with the data delivery of business audit and the result being exported to the interactive voice response module;
(5), interactive voice response module is carried out hyperchannel automatic flow control and treatment to the voice demand collection in the interactive voice flow process; And utilize its literary composition language modular converter to enter prosody modeling, text analyzing, phonetic synthesis, literary composition language correction process and realize interactive voice demand response result's flow processing is finished in the processing that text receives, voice shift, result's calibration, result send voice.
The present invention is owing to adopt mutual deduction system of man-machine language and the implementation method of being made up of ASR sound identification module, Service Processing Module, interactive voice response module, message control module and database combination, utilize acoustic feature, phonetic feature principle by sound identification module, utilize computer technology and logic of language as instrument processed voice identification process, repeatedly unclog and readjust and obtain voice identification result by language vocabulary being intersected interaction, solve the deficiency and the difficult point of existing speech recognition technology; Utilize language vocabulary sentence pattern database data to handle the text that obtains correct result is provided, solved existing phonetically similar word in the civilian language switch technology, wrongly written character malapropism technological difficulties; The automatic flow control function of utilizing the interactive voice response module that voice demand in the interactive voice deduction system is gathered solves voice demand collection hyperchannel automatic flow control in man machine language's interaction flow; According to data, physics, computer science, philosophy, the science of law, linguistics etc. as application foundation, utilize logic of language as instrument, make an explanation at language type, semanteme, pragmatic and analyze to submit the result to, solve language understanding and analyze the response result technological difficulties; Service distribution module by message control module is submitted to language historical data storehouse and Service Database matching treatment with the data result distribution of obtaining, be submitted to the business audit resume module, the business audit module will be submitted to professional output module according to the result that the audit of linguistic interpretation flow process is handled, finish the output result by professional output module function, solve the technological difficulties of matching treatment as a result of man machine language's interaction demand response.
Describe realization of the present invention in detail below in conjunction with accompanying drawing.
Description of drawings
Fig. 1 is realization flow figure of the present invention.
Embodiment
As shown in Figure 1, the mutual deduction system of man-machine language of the present invention, comprise voice acquisition module ASR sound identification module, Service Processing Module, the interactive voice response module that includes literary composition language modular converter, message control module and database, wherein, voice acquisition module is to gather multichannel people's language tone signal and it is delivered to ASR voice recognition processing module, described voice recognition processing module is utilized computer technology and logic of language, acoustic feature principle and phonetic feature principle are as instrument processed voice identification process, repeatedly unclog and readjust and obtain voice identification result and this voice identification result is delivered to Service Processing Module by language vocabulary being intersected interaction, Service Processing Module utilizes logic of language as instrument, according to mathematics, physics, computer science, philosophy, the science of law, linguistics etc. are as application foundation, at language shape, semantic, pragmatic makes an explanation and distributes and submit to the result to message control module, message control module will carry out the processing of professional control flow from the data result that Service Processing Module obtains, and described message control module comprises traffic assignments, business audit, professional output module, wherein, described service distribution module is according to the data result that obtains in the Service Processing Module, distributes to be submitted to language historical data storehouse and to mate the data result that is obtained with Service Database and carry out business audit; Described professional output module is the literary composition language modular converter of giving professional output module with the data delivery of business audit and the result being exported to the interactive voice response module, the interactive voice response module is carried out hyperchannel automatic flow control and treatment to the voice demand collection in the interactive voice flow process, and utilize its literary composition language modular converter to enter prosody modeling, text analyzing, phonetic synthesis, literary composition language correction process and realize interactive voice demand response result's flow processing is finished in the processing that text receives, voice shift, result's calibration, result send voice.
Wherein above-mentioned database comprises the phonetic feature storehouse of being made up of voice messaging storehouse and language vocabulary sentence pattern storehouse, language historical data storehouse, Service Database, wherein said phonetic feature storehouse is to utilize phonetic feature modeling image data to finish, described voice messaging storehouse is to utilize the man-machine interaction language message to be stored into the language message storehouse after handling audit, described language vocabulary sentence pattern storehouse is to utilize language vocabulary, the dialect of the multiple popular constant of sentence pattern and collection and the data of phonetic synthesis compilation thereof are finished, described language historical data storehouse comprises: the mathematics historical data, the philosophy historical data, the physics historical data, the chemistry historical data, multidisciplinary historical data such as language historical data, reference book compilation is finished, and described Service Database comprises that the data compilation of business demand collection finishes.
The mutual deduction system of described man-machine language of the present invention carries out the intelligent implementation method of man-machine language interaction demand response, may further comprise the steps:
(1), voice acquisition module is delivered to ASR voice recognition processing module with the hyperchannel people language tone signal that collects;
(2), ASR voice recognition processing module utilizes computer technology and logic of language, acoustic feature principle and phonetic feature principle as instrument processed voice identification process, repeatedly unclog and readjust and obtain voice identification result and this voice identification result is delivered to Service Processing Module by language vocabulary being intersected interaction;
(3), Service Processing Module utilizes logic of language as instrument, as application foundation, at distributions that make an explanation of language shape, semanteme, pragmatic, the result is to message control module in submission according to mathematics, physics, computer science, philosophy, the science of law, linguistics etc.;
(4), message control module will carry out the processing of professional control flow from the data result that Service Processing Module obtains, and described message control module comprises traffic assignments, business audit, professional output module, wherein, described service distribution module is according to the data result that obtains in the Service Processing Module, distributes to be submitted to language historical data storehouse and to mate the data result that is obtained with Service Database and carry out business audit; Described professional output module is the literary composition language modular converter of giving professional output module with the data delivery of business audit and the result being exported to the interactive voice response module;
(5), interactive voice response module is carried out hyperchannel automatic flow control and treatment to the voice demand collection in the interactive voice flow process; And utilize its literary composition language modular converter to enter prosody modeling, text analyzing, phonetic synthesis, literary composition language correction process and realize interactive voice demand response result's flow processing is finished in the processing that text receives, voice shift, result's calibration, result send voice.
Claims (3)
1, the mutual deduction system of a kind of man-machine language, it is characterized in that comprising voice acquisition module ASR sound identification module, Service Processing Module, the interactive voice response module that includes literary composition language modular converter, message control module and database, wherein, voice acquisition module is to gather multichannel people's language tone signal and it is delivered to ASR voice recognition processing module, described voice recognition processing module is utilized computer technology and logic of language, acoustic feature principle and phonetic feature principle are as instrument processed voice identification process, repeatedly unclog and readjust and obtain voice identification result and this voice identification result is delivered to Service Processing Module by language vocabulary being intersected interaction, Service Processing Module utilizes logic of language as instrument, according to mathematics, physics, computer science, philosophy, the science of law, linguistics etc. are as application foundation, at language shape, semantic, pragmatic enters assignment interpretation and submits to the result to message control module, message control module will enter the processing of professional control flow from the data result that Service Processing Module obtains, and described message control module comprises traffic assignments, business audit, professional output module, wherein, described service distribution module is according to the data result that obtains in the Service Processing Module, distributes to be submitted to language historical data storehouse and to mate the data result that is obtained with Service Database and enter business audit; Described professional output module is the literary composition language modular converter of giving professional output module with the data delivery of business audit and the result being exported to the interactive voice response module, the interactive voice response module enters the automatic flow control and treatment to the voice demand collection in the interactive voice flow process, and utilize its literary composition language modular converter to enter prosody modeling, text analyzing, phonetic synthesis, literary composition language correction process and realize interactive voice demand response result's flow processing is finished in the processing that text receives, voice shift, result's calibration, result send voice.
2, the mutual deduction system of man-machine language according to claim 1, it is characterized in that above-mentioned database comprises the phonetic feature storehouse of being made up of voice messaging storehouse and language vocabulary sentence pattern storehouse, language historical data storehouse, Service Database, wherein said phonetic feature storehouse is to utilize phonetic feature modeling image data to finish, described voice messaging storehouse is to utilize the man-machine interaction language message to be stored into the language message storehouse after handling audit, described language vocabulary sentence pattern storehouse is to utilize language vocabulary, the dialect of the multiple popular constant of sentence pattern and collection and the data of phonetic synthesis compilation thereof are finished, described language historical data storehouse comprises: the mathematics historical data, the philosophy historical data, the physics historical data, the chemistry historical data, multidisciplinary historical data such as language historical data, reference book compilation is finished, and described Service Database comprises that the data compilation of business demand collection finishes.
3, the mutual deduction system of man-machine language according to claim 1 carries out the intelligent implementation method of man-machine language interaction demand response, it is characterized in that may further comprise the steps:
(1), voice acquisition module is delivered to ASR voice recognition processing module with the hyperchannel people language tone signal that collects;
(2), ASR voice recognition processing module utilizes computer technology and logic of language, acoustic feature principle and phonetic feature principle as instrument processed voice identification process, repeatedly unclog and readjust and obtain voice identification result and this voice identification result is delivered to Service Processing Module by language vocabulary being intersected interaction;
(3), Service Processing Module utilizes logic of language as instrument, as application foundation, at distributions that make an explanation of language shape, semanteme, pragmatic, the result is to message control module in submission according to mathematics, physics, computer science, philosophy, the science of law, linguistics etc.;
(4), message control module will enter the processing of professional control flow from the data result that Service Processing Module obtains, and described message control module comprises traffic assignments, business audit, professional output module, wherein, described service distribution module is according to the data result that obtains in the Service Processing Module, distributes to be submitted to language historical data storehouse and to mate the data result that is obtained with Service Database and carry out business audit; Described professional output module is the literary composition language modular converter of giving professional output module with the data delivery of business audit and the result being exported to the interactive voice response module;
(5), interactive voice response module is carried out hyperchannel automatic flow control and treatment to the voice demand collection in the interactive voice flow process; And utilize its literary composition language modular converter to enter prosody modeling, text analyzing, phonetic synthesis, literary composition language correction process and realize interactive voice demand response result's flow processing is finished in the processing that text receives, voice shift, result's calibration, result send voice.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008102206499A CN101488342A (en) | 2008-12-31 | 2008-12-31 | Human-machine language interaction deduction system and intelligent implementing method for human-machine language interaction demand response |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008102206499A CN101488342A (en) | 2008-12-31 | 2008-12-31 | Human-machine language interaction deduction system and intelligent implementing method for human-machine language interaction demand response |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101488342A true CN101488342A (en) | 2009-07-22 |
Family
ID=40891195
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2008102206499A Pending CN101488342A (en) | 2008-12-31 | 2008-12-31 | Human-machine language interaction deduction system and intelligent implementing method for human-machine language interaction demand response |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101488342A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102833633A (en) * | 2012-09-04 | 2012-12-19 | 深圳创维－Rgb电子有限公司 | System and method for controlling television voice |
| CN104464731A (en) * | 2013-09-20 | 2015-03-25 | 株式会社东芝 | Data collection device, method, voice talking device and method |
| CN106228983A (en) * | 2016-08-23 | 2016-12-14 | 北京谛听机器人科技有限公司 | Scene process method and system during a kind of man-machine natural language is mutual |
| CN106663426A (en) * | 2014-07-03 | 2017-05-10 | 微软技术许可有限责任公司 | Generating computer responses to social conversational inputs |
| CN106851478A (en) * | 2017-02-10 | 2017-06-13 | 深圳市笨笨机器人有限公司 | Multi-channel information processing method and system |
| CN107886938A (en) * | 2016-09-29 | 2018-04-06 | 中国科学院深圳先进技术研究院 | Virtual reality guides hypnosis method of speech processing and device |
| CN108491517A (en) * | 2018-03-22 | 2018-09-04 | 青岛农业大学 | A kind of region agricultural information service speech polling terminal |
| US10909969B2 (en) | 2015-01-03 | 2021-02-02 | Microsoft Technology Licensing, Llc | Generation of language understanding systems and methods |
-
2008
- 2008-12-31 CN CNA2008102206499A patent/CN101488342A/en active Pending
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102833633A (en) * | 2012-09-04 | 2012-12-19 | 深圳创维－Rgb电子有限公司 | System and method for controlling television voice |
| CN104464731A (en) * | 2013-09-20 | 2015-03-25 | 株式会社东芝 | Data collection device, method, voice talking device and method |
| CN106663426A (en) * | 2014-07-03 | 2017-05-10 | 微软技术许可有限责任公司 | Generating computer responses to social conversational inputs |
| US10909969B2 (en) | 2015-01-03 | 2021-02-02 | Microsoft Technology Licensing, Llc | Generation of language understanding systems and methods |
| CN106228983A (en) * | 2016-08-23 | 2016-12-14 | 北京谛听机器人科技有限公司 | Scene process method and system during a kind of man-machine natural language is mutual |
| CN106228983B (en) * | 2016-08-23 | 2018-08-24 | 北京谛听机器人科技有限公司 | A kind of scene process method and system in man-machine natural language interaction |
| CN107886938A (en) * | 2016-09-29 | 2018-04-06 | 中国科学院深圳先进技术研究院 | Virtual reality guides hypnosis method of speech processing and device |
| CN107886938B (en) * | 2016-09-29 | 2020-11-17 | 中国科学院深圳先进技术研究院 | Virtual reality guidance hypnosis voice processing method and device |
| CN106851478A (en) * | 2017-02-10 | 2017-06-13 | 深圳市笨笨机器人有限公司 | Multi-channel information processing method and system |
| CN108491517A (en) * | 2018-03-22 | 2018-09-04 | 青岛农业大学 | A kind of region agricultural information service speech polling terminal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101488342A (en) | Human-machine language interaction deduction system and intelligent implementing method for human-machine language interaction demand response | |
| US8498857B2 (en) | System and method for rapid prototyping of existing speech recognition solutions in different languages | |
| TWI276046B (en) | Distributed language processing system and method of transmitting medium information therefore | |
| CN103680498A (en) | Speech recognition method and speech recognition equipment | |
| WO2002033542A3 (en) | Software development systems and methods | |
| CN103003876A (en) | Modification of speech quality in conversations over voice channels | |
| Jimerson et al. | ASR for documenting acutely under-resourced indigenous languages | |
| CN108184032B (en) | Service method and device of customer service system | |
| CN107274889A (en) | A kind of method and device according to speech production business paper | |
| CN108763338A (en) | A kind of News Collection&Edit System based on power industry | |
| Sakti et al. | Development of Indonesian large vocabulary continuous speech recognition system within A-STAR project | |
| CN1901041B (en) | Voice dictionary forming method and voice identifying system and its method | |
| CN108446278A (en) | A kind of semantic understanding system and method based on natural language | |
| CN110781649A (en) | Subtitle editing method and device, computer storage medium and electronic equipment | |
| Matoušek et al. | Building of a speech corpus optimised for unit selection TTS synthesis | |
| CN110909879A (en) | Auto-regressive neural network disambiguation model, training and using method, device and system | |
| CN106356054A (en) | Method and system for collecting information of agricultural products based on voice recognition | |
| CN1333501A (en) | Dynamic Chinese speech synthesizing method | |
| CN112509550A (en) | Speech synthesis model training method, speech synthesis device and electronic equipment | |
| CN111914078A (en) | Data processing method and device | |
| CN1032391C (en) | Chinese character-phonetics transfer method and system edited based on waveform | |
| Eckert et al. | Real users behave weird-Experiences made collecting large human-machine-dialog corpora | |
| Callejas et al. | Implementing modular dialogue systems: A case of study | |
| CN101958118A (en) | Implement the system and method for speech recognition dictionary effectively | |
| CN110728980A (en) | Intelligent service bus system based on voice and conversation robot |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Open date: 20090722 |