CN1694162A - Phonetic recognition analysing system and service method - Google Patents
Phonetic recognition analysing system and service method Download PDFInfo
- Publication number
- CN1694162A CN1694162A CNA2005100461691A CN200510046169A CN1694162A CN 1694162 A CN1694162 A CN 1694162A CN A2005100461691 A CNA2005100461691 A CN A2005100461691A CN 200510046169 A CN200510046169 A CN 200510046169A CN 1694162 A CN1694162 A CN 1694162A
- Authority
- CN
- China
- Prior art keywords
- sound
- voice
- degree
- person
- servicing
- 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
Landscapes
- Telephonic Communication Services (AREA)
Abstract
The invention relates to a voice identify method, especially relate to the device of the voice identify system and the method of service. It is characterized that it pick-up the frequency of voice at the conversation, and based on the voice affective degree and affinity of the voice technically, identifying and analysising voice a based on the sensibility science area. The voice affective degree can know the person's character and psychology status of the sounder according to the people soniferous tone and tonilaty and show the real feeling of sounder. Through the INTERNET and correspondence termination device PDA, data TV, mobile phone, net phone, fixed phone, we can realize the transmit of the analytic result and advice as well as the whole service from the system. The effect of the invention is to realize the voice analysis which is based on the field of sensibility science.
Description
Technical field
The invention belongs to the computer communication technology field, relate to audio recognition method, specially refer to speech recognition system equipment and method of servicing.
Background technology
Traditional speech recognition technology is exactly that the language of personnel selection removes commander and control machine, allows machine " understand " people's language, and goes to finish various tasks according to its instruction.
Traditional speech recognition system is divided into three parts haply:
(1) phonetic feature extracts: its objective is to extract time dependent phonetic feature sequence from speech waveform.
(2) acoustic model and pattern match (recognizer): acoustic model produces the phonetic feature that obtains usually by learning algorithm.Phonetic feature with input when identification mates and compares with acoustics model (pattern), obtains best recognition result.
(3) language model and Language Processing: language model comprises grammer network that is made of voice command recognition or the language model that is made of statistical method, and Language Processing can be carried out grammer, semantic analysis.To little vocabulary speech recognition system, often do not need the Language Processing part.
Traditional speech processing system has just reflected an aspect of information, and has ignored corresponding with knowledge world and have the perceptual Scientific World of status of equal importance, and this also is an important component part of information processing in fact.
Traditional speech processing system pattern is simple, and the fixing language model of simple basis goes out people's language by machine recognition.Because this simple recognition mode has caused the unicity of application facet, normally is applied to information inquiry, paging centers, call forwarding center; The financial field is as Home Banking or stock information inquiry; The tourist service field, as aircraft booking or hotel reservation, etc.But along with the fast development of science and technology, the improving constantly of human's demand, speech recognition system should have broader applications, combines with the analysis of emotion aspect, embodies human language true essence truly.
The deficiency of above-mentioned speech recognition system and method can be overcome by a kind of brand-new speech recognition system and method.
Summary of the invention
The purpose of this invention is to provide a kind of is technical basis according to the frequency of extracting sound in person to person's communication process with sound emotion degree and sound affinity degree, realizes identification and analysis based on perceptual scientific domain, the speech recognition system that draws thus.
It is the identification and the analysis of the perceptual scientific domain of technical basis with sound emotion degree and sound affinity degree that another object of the present invention provides a kind of.
Another object of the present invention provides a kind of score value suggestion based on the human psychology state.
Another object of the present invention provides the speech recognition system that a kind of INTERNET and communication facilities terminal combine.
Technical scheme of the present invention is to converse via speech recognition system in human use's communication facilities terminal, system is the technical basis analysis mankind's sound and provides the score value suggestion with the affinity and the emotion degree of sound in this process, finally be connected user with information interaction between speech analysis system by INTERNET with communication terminal device (as: PDA, Digital Television, mobile phone, the networking telephone, fixed line etc.) with mail server, realize the reception and registration and the full-range service of analysis result and suggestion by short message server, voice server, socket server.
Sound affinity degree can be interpreted as: human sound has intrinsic separately frequency, promptly by the audio frequency that tone and musical note constituted of uniqueness.Its medium pitch, i.e. range, just the width of sound can be judged this person's tonequality according to the height of speaker's range, and then understands his personality; And musical note promptly is the fluctuation degree of a personal voice, by analyzing a people's musical note, can grasp speaker's psychological condition at that time.Extract their audio frequency from 1000 pairs of philtrums, find out a reference value according to the distribution of their audio frequency, when judging two people's sound affinity, just their sound is analyzed, thereby find out gap value, then gap value and reference value being compared to draw a number percent, and this number percent is exactly the affinity of sound.Generally, the sound affinity between lineal relative and spouse is about 90%, and the sound affinity between friend is about 80%, and when two people met for the first time, their sound affinity was generally about 70%.
The emotion degree of sound can be interpreted as: human information by the sound statement is divided two kinds, and a kind of is the communicate information that two people are passed on when exchanging, and a kind of is the overseas publicity information of self-mood.Can show speaker's mood really by the low frequency sounding that lung directly drives, this frequency is the most basic frequency of the factor of removing all kinds of obstacles, and also is the base frequency of decision speaker sound emotion degree.Allow 2000 people read same passage, and gather their low frequency sounding respectively, thereby obtain a reference value, allow the 2001st people also read same literal then, and gather his low frequency sounding, then his audible frequency value and reference value are compared and obtain a number percent, if this number percent is 50%, so conclusion be exactly the 2001st people affective characteristics just and the 1000th people's affective characteristics match.Generally, the sound emotion degree the when newsreader broadcasts is about 55%, if when a people is happy, his sound emotion degree should be about 70%, and when he laughed heartily, sound emotion degree can rise to 80%.
Provide following six aspect services according to above-mentioned technical basis native system:
1, is in unrequited love with successfully degree test
Be in unrequited love with others the calling party can by we system with conversed by unrequited love person, system can analyze the other side's sound and provide the score value suggestion, make unrequited love person can know the other side's regard, so both can avoid some unnecessary embarrassments, can solve unrequited love person's worries again.
2, love sweetness degree test
Calling party in the love can converse by the lover of native system with oneself, and system can analyze their sound and provide the score value suggestion, makes its thought timely the other side of assurance of each stage of love dynamic, makes reply.
3, the harmonious degree test of marriage
Married calling party can converse by native system and spouse, and system can analyze their sound and provide score value suggestion, makes its emotion degree of understanding both sides constantly, in time examines oneself or reminds the other side.
4, friendship weight test
The calling party can converse by native system and its friend, and system can analyze and provide score value behind their sound and advise, and the calling party can understand the view of friend to himself at any time like this.
5, father and mother's level of understanding test
Calling party as children can converse by native system and its father and mother, and system can analyze and provide score value to their sound, and children just can understand the attitude of father and mother's idea to it at any time like this, help two generations's communication.
6, children's degree of rebel test
Calling party as father and mother can converse by native system and its children, and system can analyze its sound and provide the score value suggestion, and father and mother just can come into child's heart like this, understand their idea, reduce the generation of generation gap.
In an example, for the test of being in unrequited love with successful degree, provide in the speech recognition system of the present invention based on the sound collection data of being in unrequited love with successful degree, the user according to system suggestion converses with tested user by short message terminal equipment (mobile phone, landline telephone, the networking telephone etc.), provide the score value suggestion after system's process sound collection, coupling, the analysis, inform to the user by the communications conduit that the user is used.
Effect of the present invention and benefit are: realized the speech analysis based on perceptual scientific domain.
Description of drawings
Fig. 1 is the central database structural representation of speech recognition system.
Among the figure: 1 network, 2 intermediary systems, 3 user's terminals, 4 short message servers, 5 socket servers, 6 mail servers, 7 database servers, 8 audio databases, 9 customer data bases, 10 score value databases.
Fig. 2 is that test is prepared and the reception result synoptic diagram.
Among the figure: whether 11 test, 12 select method of service, the test of 13 affinity, 14 speech analysises, 15 note analyses, 16 select test-types, 17 input passwords, and whether 18 passwords are correct, 19 input handset numbers, whether 20 phone numbers are correct, and 21 receive the note analysis result, and 22 listen to result of voice analysis.
Fig. 3 is the affinity test pattern.
Among the figure: 23 are in unrequited love with successfully degree, 24 love sweetness degree, and the harmonious degree of 25 marriages, 26 friendship weights, 27 father and mother's levels of understanding, 28 children's degree of rebel, 29 receiving system passwords, 30 receiving system passwords whether again, 31 dial called number, and 32 converse 33 on-hooks.
Embodiment
Below in conjunction with technical scheme and accompanying drawing, describe most preferred embodiment of the present invention in detail.
Below in conjunction with the description of the drawings is description as currently preferred embodiments of the invention, and unique form of not representing the present invention to be configured and/or to use.Provided in conjunction with illustrated embodiment during description and be used for structure and operate function of the present invention and sequence of steps, yet, should be appreciated that function identical or that be equal to and also can finish in proper order by the different enforcement that comprises within the spirit and scope of the present invention.
Speech analysis system comprises intermediary system 2 and the user test terminal 3 by the all-network connection of various forms such as network 1, telephone network, internet, wireless communication networks.User test terminal 3 is to connect the signal equipment that intermediary system 2 is tested by network 1.And the tester utilizes PC communication, internet, wireless network, telephone network etc., simply and promptly visits intermediary system 2.Be connected with network 1 by this tester's terminal 3, the intermediary system 2 that the tester carries out tone testing is to comprise the short message server 4 that passes through network 1 visit and message exchange, the various information that socket server 5 and mail server 6 etc. are passed on, testee's voice messaging flows to and carries out data processing by storing a lot of database of information servers 7 in the test process.Database server 7 comprises storage and handles the audio database 8 of wave audio, the customer data base net 9 of storage user information and inventory analysis result's score value database 10.Network interface server 5 is by network 1 wireless network access tester terminal 3 particularly, the various information that provides by database server 7 offers tester's terminal 3 after handling according to rule transformation, offer database server 7 after the information that transmits by network 1 acceptance test person's terminal 3, comprise database linkage that the information of the webserver or other system transmits etc.Mail server 6 carries out the transmission of Email by network 1 and survey envoy terminal 3.
Though the present invention illustrates that with reference to real system the present invention not only is limited in above-described content.
Claims (4)
1, a kind of phonetic recognition analysing system and method for servicing, it is characterized in that in person to person's communication process, extracting human sound frequency, with sound emotion degree and sound affinity is technical basis, carry out speech analysis and identification based on the perceptual scientific domain of the mankind, the score value suggestion is provided simultaneously, by INTERNET and communication terminal device PDA, Digital Television, mobile phone, the networking telephone, landline telephone, connect the information interaction between user and speech recognition system by short message server, voice server, socket server and mail server.
2, according to right 1 described a kind of phonetic recognition analysing system and method for servicing, it is characterized in that in person to person's communication process, directly extracting human sound frequency, and be the identification and the analysis of the perceptual scientific domain of technical basis with sound emotion degree and sound affinity degree by system.
3,, it is characterized in that according to the suggestion of systematic analysis link score value according to right 1 described a kind of phonetic recognition analysing system and method for servicing.
4, according to right 1 described a kind of phonetic recognition analysing system and method for servicing, it is characterized in that connecting the information interaction between user and speech recognition system by INTERNET and communication terminal device PDA, Digital Television, mobile phone, the networking telephone, landline telephone.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2005100461691A CN1694162A (en) | 2005-03-31 | 2005-03-31 | Phonetic recognition analysing system and service method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2005100461691A CN1694162A (en) | 2005-03-31 | 2005-03-31 | Phonetic recognition analysing system and service method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1694162A true CN1694162A (en) | 2005-11-09 |
Family
ID=35353098
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2005100461691A Pending CN1694162A (en) | 2005-03-31 | 2005-03-31 | Phonetic recognition analysing system and service method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1694162A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| ES2306561A1 (en) * | 2005-12-30 | 2008-11-01 | France Telecom España, S.A. | Method for measuring the degree of affinity between people through voice biometry in mobile devices (Machine-translation by Google Translate, not legally binding) |
| CN101645961A (en) * | 2008-08-06 | 2010-02-10 | 深圳富泰宏精密工业有限公司 | Mobilephone and method for achieving caller emotion identification |
| CN101867742A (en) * | 2010-05-21 | 2010-10-20 | 中山大学 | Television system based on sound control |
| CN103440864A (en) * | 2013-07-31 | 2013-12-11 | 湖南大学 | Personality characteristic forecasting method based on voices |
| CN104900007A (en) * | 2015-06-19 | 2015-09-09 | 四川分享微联科技有限公司 | Monitoring watch triggering wireless alarm based on voice |
| CN105141587A (en) * | 2015-08-04 | 2015-12-09 | 广东小天才科技有限公司 | Virtual doll interaction method and device |
| CN105206269A (en) * | 2015-08-14 | 2015-12-30 | 百度在线网络技术（北京）有限公司 | Voice processing method and device |
| CN109377540A (en) * | 2018-09-30 | 2019-02-22 | 网易（杭州）网络有限公司 | Synthetic method, device, storage medium, processor and the terminal of FA Facial Animation |
-
2005
- 2005-03-31 CN CNA2005100461691A patent/CN1694162A/en active Pending
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| ES2306561A1 (en) * | 2005-12-30 | 2008-11-01 | France Telecom España, S.A. | Method for measuring the degree of affinity between people through voice biometry in mobile devices (Machine-translation by Google Translate, not legally binding) |
| CN101645961A (en) * | 2008-08-06 | 2010-02-10 | 深圳富泰宏精密工业有限公司 | Mobilephone and method for achieving caller emotion identification |
| CN101867742A (en) * | 2010-05-21 | 2010-10-20 | 中山大学 | Television system based on sound control |
| CN103440864A (en) * | 2013-07-31 | 2013-12-11 | 湖南大学 | Personality characteristic forecasting method based on voices |
| CN104900007A (en) * | 2015-06-19 | 2015-09-09 | 四川分享微联科技有限公司 | Monitoring watch triggering wireless alarm based on voice |
| CN105141587A (en) * | 2015-08-04 | 2015-12-09 | 广东小天才科技有限公司 | Virtual doll interaction method and device |
| CN105141587B (en) * | 2015-08-04 | 2019-01-01 | 广东小天才科技有限公司 | A kind of virtual puppet interactive approach and device |
| CN105206269A (en) * | 2015-08-14 | 2015-12-30 | 百度在线网络技术（北京）有限公司 | Voice processing method and device |
| CN109377540A (en) * | 2018-09-30 | 2019-02-22 | 网易（杭州）网络有限公司 | Synthetic method, device, storage medium, processor and the terminal of FA Facial Animation |
| CN109377540B (en) * | 2018-09-30 | 2023-12-19 | 网易（杭州）网络有限公司 | Method and device for synthesizing facial animation, storage medium, processor and terminal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109817213B (en) | Method, device and equipment for performing voice recognition on self-adaptive language | |
| CN1694162A (en) | Phonetic recognition analysing system and service method | |
| CN100351899C (en) | Intermediary for speech processing in network environments | |
| US10102847B2 (en) | Automated learning for speech-based applications | |
| CN101297355B (en) | Systems and methods for responding to natural language speech utterance | |
| CN107818798A (en) | Customer service quality evaluating method, device, equipment and storage medium | |
| CN105744090A (en) | Voice information processing method and device | |
| CN109325091B (en) | Method, device, equipment and medium for updating attribute information of interest points | |
| US20120262296A1 (en) | User intent analysis extent of speaker intent analysis system | |
| CN109949071A (en) | Products Show method, apparatus, equipment and medium based on voice mood analysis | |
| KR20120038000A (en) | Method and system for determining the topic of a conversation and obtaining and presenting related content | |
| CN111862942B (en) | Method and system for training mixed speech recognition model of Mandarin and Sichuan | |
| CN110633912A (en) | Method and system for monitoring service quality of service personnel | |
| CN115602165B (en) | Digital employee intelligent system based on financial system | |
| CN111489765A (en) | Telephone traffic service quality inspection method based on intelligent voice technology | |
| CN109670055A (en) | A kind of multi-medium data checking method, device, equipment and storage medium | |
| CN115798518B (en) | Model training method, device, equipment and medium | |
| Abdullah et al. | Paralinguistic speech processing: An overview | |
| KR102171658B1 (en) | Crowd transcription apparatus, and control method thereof | |
| CN110458599A (en) | Test method, test device and Related product | |
| EP1382032B1 (en) | Web-based speech recognition with scripting and semantic objects | |
| CN111949777A (en) | Intelligent voice conversation method and device based on crowd classification and electronic equipment | |
| CN111858884A (en) | Method and system for robot to learn real person deep dialogue content | |
| KR20030033890A (en) | Cyber secretary system using analysis of user's voice information and method thereof | |
| CN112116165B (en) | Business performance determining method and device |
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 |