CN102222502A - Effective way for voice verification by Chinese text-prompted mode - Google Patents
Effective way for voice verification by Chinese text-prompted mode Download PDFInfo
- Publication number
- CN102222502A CN102222502A CN201110125644XA CN201110125644A CN102222502A CN 102222502 A CN102222502 A CN 102222502A CN 201110125644X A CN201110125644X A CN 201110125644XA CN 201110125644 A CN201110125644 A CN 201110125644A CN 102222502 A CN102222502 A CN 102222502A
- Authority
- CN
- China
- Prior art keywords
- vocal print
- user
- voice
- random
- text
- 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
- Document Processing Apparatus (AREA)
Abstract
The invention relates to an effective way for voice verification by a Chinese text-prompted mode. The effective way is completed by the following steps that: a voice password verification system of a text-prompted mode of a text-dependent type is used, wherein the voice password verification system of the text-prompted mode of the text-dependent type comprises a voice enrollment system, a random type password generator, and a voice verification system; when voice enrollment is carried out, a user speaks zero, one, until nine as well as relevant segmentation words; after the user finished, the voice enrollment system establishes and stores voice materials of the user; when voice verification is carried out, the random type password generator selects an effective date, time, a sum, or a house number, and the user is asked to read the selected effective date, the selected time, the selected sum, or the selected house number; the voice verification system embodies after-reading voice information of the user and compares the information with the similar probability with minimum limitation; if the result is larger that the similar probability with minimum limitation, the information is passed; if not, the information is not allowed to be passed. According to a technical scheme of the invention, when the user speaks for a long time during the voice verification, the accuracy of the voice verification is high; Besides, because the segmentation words including year, month, and date are employed, a long random numeric string is separated into three short parts, so that the random numeric string can be easy to be remembered by the user.
Description
Technical field
The present invention relates to biological characteristic and distinguish the field, be specifically related to the effective means that the vocal print checking pointed out at random in a kind of Chinese.
Background technology
The most frequently used personal verification mode is a password at present, and in this authentication mode, the user must keep password firmly in mind, for example user's account No. password, credit card password, network entry account number cipher etc.Because to the requirement difference of password, the user must learn a plurality of different passwords by heart in need not using, very inconvenience, and, in case password is stolen, will cause very big loss to the user.
The development of biological characteristic recognition techniques makes the authentication mode obtain great progress, for example face identification, identification of fingerprint, palmmprint identification, sound-groove identification, the identification of eyeball iris and DNA (deoxyribonucleic acid) (DNA) ordering pairing identification etc.If during based on the framework cost of considering identification technique and practical operation, can find to be fit to selected purposes as the personal identification identification by relative other identification modes of vocal print checking identification, also be identification mode comparatively commonly used.
Sound-groove identification is to compare by the acoustic information that will import and the vocal print feature of storage, thereby confirms the verification mode of user's identity.This vocal print is verified the difference on the different stages such as individual difference that utilize between the individual difference on the vocal organs between the speaker, the individual difference between the pronunciation channel, pronunciation custom, and COMPREHENSIVE CALCULATING machine digital information processing is realized the identification of user's identity.
The vocal print checking has registration (Voice Enrollment) and checking (VoiceVerification) two links, the correlativity of the content of saying during according to when checking and registration, vocal print verification technique are divided into text (Text-Dependent) uncorrelated with text (Text-Independent) two big classes of being correlated with; Whether fully and consistent during registration its Chinese version related side formula the time speaks content according to checking again, is divided into classics (Classic) pattern and points out (Text-Prompted) pattern at random.
The speech data that need gather when the uncorrelated mode of text is registered is more, and the user voice data that needs during checking is also more, and the process of simultaneous verification is also longer, and the backstage that usually is applied in user and the manual position communication process is detected.This mode is not suitable for real-time and requires situation high, that user's vocal print password is short.
The classical mode that text is relevant exists the user to speak content by the security risk of resetting after being recorded owing to must say same content at every turn.Vocal print verifies that manufacturer provides recording playback measuring ability, but detection is the content that the user said on the history in its system, and the playback measuring ability has two restrictions greatly: the one, and the limit of reservation historical data, the 2nd, can't detect the content of recording by other channels.Therefore there is security risk in the classical mode that text is relevant.
The prompt modes at random that text is relevant requires the user to say certain voice content when registering, and asks the user with reading, by the randomness solution playback problem of content with randomly drawing content when checking.
The relevant existing implementation of prompt modes at random of text is a numeric string.To be applied as example abroad:
1). ask the user to say zero, one during registration, please the user say at random during checking as " six five nine two " to nine
2). registration is to ask the user to say at random four numerals (4-DigitChallenge/Response), as " six five nine two ", asks the user to say in these four various permutation and combination of numeral one as " nine five six two " during checking.
Because Chinese character syllable is short, it is bad that this implementation is applied in the Chinese effect, causes vocal print authentication error rate FAR/FRR (False Accept Rate/False Reject Rate) to increase, and can't satisfy the demand for security of vocal print checking.
If increase the random number word string, can improve the error rate of vocal print checking, but the user is difficult for memory and with reading a string random number word string that has no rule, cause user experience to descend.
Sum up it, there is following shortcoming in existing vocal print verification technique:
The speech data of the registration of-text-independent pattern, checking is many, and proof procedure is slow, is not suitable for real-time, the password authentification of user's vocal print
There is the security risk of recording playback in the classical mode that-text is relevant
Prompt modes at random (4DCR) the playback risk probability that-text is relevant is n/10000
The prompt modes at random that-text is relevant exists the vocal print checking accuracy rate that causes because of the Chinese syllable short-range missile to descend (words that the accidental validation content is too short), or user experience decline (words that the accidental validation content is long).
Summary of the invention
The technical problem to be solved in the present invention is: at be the relevant prompt modes at random of text, purpose is to provide a kind of Chinese to point out the effective means of vocal print checking at random, solves this pattern accuracy rate in Chinese application and descends the problem that user experience descends.
The technical solution adopted for the present invention to solve the technical problems is: the effective means of vocal print checking pointed out at random in a kind of Chinese, it is characterized in that may further comprise the steps:
1.. use the relevant vocal print password validation system of pointing out at random of a text, described text is relevant points out the vocal print password validation system at random, comprises a vocal print Accreditation System, a random mode password generator and a vocal print verification system;
2.. during the vocal print registration, ask the user to say that zero to nine reach the relevant speech of cutting apart;
3.. after the user finishes, the voiceprint data that described vocal print Accreditation System is set up and stored this user;
4.. during the vocal print checking, described random mode password generator is selected date of expiration, time, the amount of money or a number, asks the user with reading;
5.. described vocal print verification system is included this user with the voiceprint of reading, and makes comparisons to limiting minimum similar probability, is higher than then and passes through, and is lower than then and mustn't passes through.
Adopt date, time, the amount of money, number to verify content as vocal print.
Adopt date, Hour Minute Second, first angle branch, chamber, building, lane to verify content as vocal print.
Beneficial effect: owing to adopt technique scheme, the content that the user says during the vocal print checking is long, the accuracy rate height of vocal print checking.
Remove and cut apart speech, but the numeral of random choose has 6 to 9, even consider the codomain problem of date of expiration, its randomness is higher than 4 bit digital;
Because separation speech such as year, month, day have been arranged, long random number word string organically is divided into three short numbers, is convenient to user's memory.Especially, codomains such as year, month, day are the valid data in user's general knowledge, further facilitate user's memory.
The invention has the beneficial effects as follows: adopt the technical program, the content that the user says during the vocal print checking is long, the accuracy rate height of vocal print checking; Because separation speech such as year, month, day have been arranged, long random number word string organically is divided into three short numbers, is convenient to user's memory.
Description of drawings
Fig. 1 points out vocal print password validation system block diagram at random for text is relevant;
Fig. 2 points out the effective means specific embodiment of vocal print checking at random for a kind of Chinese of most preferred embodiment of the present invention;
Fig. 3 points out the process flow diagram of the effective means of vocal print checking at random for 21 kinds of Chinese of the embodiment of the invention.
Embodiment
In conjunction with the accompanying drawings, the present invention is further detailed explanation.These accompanying drawings are the synoptic diagram of simplification, basic structure of the present invention only is described in a schematic way, so it only show the formation relevant with the present invention.
Fig. 1 discloses the relevant vocal print password validation system block diagram of pointing out at random of text, and described text is relevant points out the vocal print password validation system at random, comprises a vocal print Accreditation System 1, a random mode password generator 2 and a vocal print verification system 3.
Fig. 2 discloses the effective means specific embodiment that the vocal print checking pointed out at random in a kind of Chinese of most preferred embodiment of the present invention, and step is as follows:
The user enters for the first time that a text is relevant points out the vocal print password validation system at random, at first enters this vocal print Accreditation System and carries out the vocal print registration, so that the voiceprint data of setting up and storing this user;
When vocal print is registered, ask the user to say zero to nine and date, Hour Minute Second, first angle is divided or do chamber, building etc., sets up voiceprint data after succeeding in registration;
During use, random mode password generator is selected a date of expiration " 201 zero year Dec 20 ", please the user with reading, the vocal print verification system is included the voiceprint that the user reads this date, and carry out similar probability to the voiceprint data of storage and calculate, limit minimum similar probability if reach, then pass through, be lower than minimum similar probability and then mustn't pass through.
Because with this date is that the Chinese character length of example random train is 11 Chinese characters, quite or be slightly more than the syllable number of four english digit, ensured the accuracy rate of vocal print checking.
Remove and cut apart speech, but the numeral of random choose has 8, consider the time of date of expiration, month has only Dec, and fate has only 30 days, these codomain problem, and its randomness is higher than 4 bit digital;
Because separation speech such as year, month, day have been arranged, long random number word string organically is divided into three short numbers, is convenient to user's memory.Especially, codomains such as year, month, day are the valid data in user's general knowledge, further facilitate user's memory.
Fig. 3 discloses the process flow diagram that the effective means of vocal print checking pointed out at random in 21 kinds of Chinese of the embodiment of the invention, and a kind of Chinese of preferred embodiment of the present invention points out the effective means step of vocal print checking as follows at random:
The user enters for the first time that a text is relevant points out the vocal print password validation system at random, at first enters this vocal print Accreditation System and carries out the vocal print registration, so that the voiceprint data of setting up and storing this user;
When vocal print is registered, ask the user to say zero to nine and date, Hour Minute Second, first angle is divided or do chamber, building etc., sets up voiceprint data after succeeding in registration;
During use, random mode password generator random choose date of expiration, time, the amount of money or a number ask the user with reading, the vocal print verification system is included user's voiceprint, and carry out similar probability to the voiceprint data of storage and calculate, limit minimum similar probability if reach, then pass through, be lower than minimum similar probability and then mustn't pass through.
More than show and described ultimate principle of the present invention and principal character and advantage of the present invention.The technician of the industry should understand; the present invention is not restricted to the described embodiments; that describes in the foregoing description and the instructions just illustrates the part effective means of the present invention in Chinese application; without departing from the spirit and scope of the present invention; the present invention is applicable to any checking content of utilizing the Chinese language feature, and these checking contents all fall in the claimed scope of the invention.The claimed scope of the present invention is defined by appending claims and equivalent thereof.
Claims (3)
1. the effective means of vocal print checking pointed out at random in a Chinese, it is characterized in that may further comprise the steps:
1.. use the relevant vocal print password validation system of pointing out at random of a text, described text is relevant points out the vocal print password validation system at random, comprises a vocal print Accreditation System, a random mode password generator and a vocal print verification system;
2.. during the vocal print registration, ask the user to say that zero to nine reach the relevant speech of cutting apart;
3.. after the user finishes, the voiceprint data that described vocal print Accreditation System is set up and stored this user;
4.. during the vocal print checking, described random mode password generator is selected date of expiration, time, the amount of money or a number, asks the user with reading;
5.. described vocal print verification system is included this user with the voiceprint of reading, and makes comparisons to limiting minimum similar probability, is higher than then and passes through, and is lower than then and mustn't passes through.
2. the effective means of vocal print checking pointed out at random in a kind of Chinese according to claim 1, it is characterized in that: adopt date, time, the amount of money, number to verify content as vocal print.
3. the effective means of vocal print checking pointed out at random in a kind of Chinese according to claim 2, it is characterized in that: adopt date, Hour Minute Second, first angle branch, chamber, building, lane to verify content as vocal print.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110125644XA CN102222502A (en) | 2011-05-16 | 2011-05-16 | Effective way for voice verification by Chinese text-prompted mode |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110125644XA CN102222502A (en) | 2011-05-16 | 2011-05-16 | Effective way for voice verification by Chinese text-prompted mode |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102222502A true CN102222502A (en) | 2011-10-19 |
Family
ID=44779036
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110125644XA Pending CN102222502A (en) | 2011-05-16 | 2011-05-16 | Effective way for voice verification by Chinese text-prompted mode |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102222502A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102496365A (en) * | 2011-11-30 | 2012-06-13 | 上海博泰悦臻电子设备制造有限公司 | User verification method and device |
| CN102694653A (en) * | 2012-04-28 | 2012-09-26 | 常州北大众志网络计算机有限公司 | Method of logging into system through speech recognition and/or facial image recognition |
| CN104036780A (en) * | 2013-03-05 | 2014-09-10 | 阿里巴巴集团控股有限公司 | Man-machine identification method and system |
| CN104660413A (en) * | 2015-01-28 | 2015-05-27 | 中国科学院数据与通信保护研究教育中心 | Voiceprint password certifying method and device |
| CN105224844A (en) * | 2014-07-01 | 2016-01-06 | 腾讯科技（深圳）有限公司 | Verification method, system and device |
| CN105575391A (en) * | 2014-10-10 | 2016-05-11 | 阿里巴巴集团控股有限公司 | Voiceprint information management method, voiceprint information management device, identity authentication method, and identity authentication system |
| CN105913850A (en) * | 2016-04-20 | 2016-08-31 | 上海交通大学 | Text related vocal print password verification method |
| TWI678635B (en) * | 2018-01-19 | 2019-12-01 | 遠傳電信股份有限公司 | Voiceprint certification method and electronic device thereof |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6496800B1 (en) * | 1999-07-07 | 2002-12-17 | Samsung Electronics Co., Ltd. | Speaker verification system and method using spoken continuous, random length digit string |
| CN1610294A (en) * | 2003-10-24 | 2005-04-27 | 阿鲁策株式会社 | Vocal print authentication system and vocal print authentication program |
| CN1905445A (en) * | 2005-07-27 | 2007-01-31 | 国际商业机器公司 | System and method of speech identification using mobile speech identification card |
| CN101753300A (en) * | 2008-12-02 | 2010-06-23 | 财团法人资讯工业策进会 | Device and method thereof for producing and verifying voice signature of message |
-
2011
- 2011-05-16 CN CN201110125644XA patent/CN102222502A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6496800B1 (en) * | 1999-07-07 | 2002-12-17 | Samsung Electronics Co., Ltd. | Speaker verification system and method using spoken continuous, random length digit string |
| CN1610294A (en) * | 2003-10-24 | 2005-04-27 | 阿鲁策株式会社 | Vocal print authentication system and vocal print authentication program |
| CN1905445A (en) * | 2005-07-27 | 2007-01-31 | 国际商业机器公司 | System and method of speech identification using mobile speech identification card |
| CN101753300A (en) * | 2008-12-02 | 2010-06-23 | 财团法人资讯工业策进会 | Device and method thereof for producing and verifying voice signature of message |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102496365A (en) * | 2011-11-30 | 2012-06-13 | 上海博泰悦臻电子设备制造有限公司 | User verification method and device |
| CN102694653A (en) * | 2012-04-28 | 2012-09-26 | 常州北大众志网络计算机有限公司 | Method of logging into system through speech recognition and/or facial image recognition |
| CN104036780A (en) * | 2013-03-05 | 2014-09-10 | 阿里巴巴集团控股有限公司 | Man-machine identification method and system |
| US9571490B2 (en) | 2013-03-05 | 2017-02-14 | Alibaba Group Holding Limited | Method and system for distinguishing humans from machines |
| CN104036780B (en) * | 2013-03-05 | 2017-05-24 | 阿里巴巴集团控股有限公司 | Man-machine identification method and system |
| CN105224844B (en) * | 2014-07-01 | 2020-01-24 | 腾讯科技（深圳）有限公司 | Verification method, system and device |
| CN105224844A (en) * | 2014-07-01 | 2016-01-06 | 腾讯科技（深圳）有限公司 | Verification method, system and device |
| CN105575391A (en) * | 2014-10-10 | 2016-05-11 | 阿里巴巴集团控股有限公司 | Voiceprint information management method, voiceprint information management device, identity authentication method, and identity authentication system |
| CN105575391B (en) * | 2014-10-10 | 2020-04-03 | 阿里巴巴集团控股有限公司 | Voiceprint information management method and device and identity authentication method and system |
| US10593334B2 (en) | 2014-10-10 | 2020-03-17 | Alibaba Group Holding Limited | Method and apparatus for generating voiceprint information comprised of reference pieces each used for authentication |
| CN104660413A (en) * | 2015-01-28 | 2015-05-27 | 中国科学院数据与通信保护研究教育中心 | Voiceprint password certifying method and device |
| CN105913850A (en) * | 2016-04-20 | 2016-08-31 | 上海交通大学 | Text related vocal print password verification method |
| CN105913850B (en) * | 2016-04-20 | 2019-05-28 | 上海交通大学 | Text correlation vocal print method of password authentication |
| TWI678635B (en) * | 2018-01-19 | 2019-12-01 | 遠傳電信股份有限公司 | Voiceprint certification method and electronic device thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102222502A (en) | Effective way for voice verification by Chinese text-prompted mode | |
| JP4463526B2 (en) | Voiceprint authentication system | |
| CN104598796B (en) | Personal identification method and system | |
| US7447632B2 (en) | Voice authentication system | |
| US8010367B2 (en) | Spoken free-form passwords for light-weight speaker verification using standard speech recognition engines | |
| US8190437B2 (en) | Speaker verification methods and apparatus | |
| CN105913850B (en) | Text correlation vocal print method of password authentication | |
| US20140081640A1 (en) | Speaker verification methods and apparatus | |
| CN102737634A (en) | Authentication method and device based on voice | |
| JPH0345417B2 (en) | ||
| CN102402985A (en) | Voiceprint authentication system for improving voiceprint identification security and implementation method thereof | |
| WO2010061194A2 (en) | Biometric identity verification | |
| WO2007001602A2 (en) | Speech recognition system for secure information | |
| CN104064189A (en) | Vocal print dynamic password modeling and verification method | |
| CN110363639B (en) | Financial management system based on artificial intelligence | |
| JP2013120540A (en) | Authentication system, registration device, authentication device, and portable medium | |
| US20220366916A1 (en) | Access control system | |
| US20070233667A1 (en) | Method and apparatus for sample categorization | |
| JP2020154496A (en) | Authentication system and authentication method | |
| JP2008158627A (en) | Personal identification device and personal identification method | |
| KR102604319B1 (en) | Speaker authentication system and method | |
| CN102360544A (en) | Method for performing voiceprint verification of Chinese through single arithmetic | |
| US8254530B2 (en) | Authenticating personal identification number (PIN) users | |
| CN109473107B (en) | Text semi-correlation voiceprint recognition method and system | |
| KR20140029990A (en) | System and method for authetificate the user using biometrics |
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: 20111019 |