CN204465555U - Based on the voiceprint authentication apparatus of time type dynamic password - Google Patents
Based on the voiceprint authentication apparatus of time type dynamic password Download PDFInfo
- Publication number
- CN204465555U CN204465555U CN201520223978.4U CN201520223978U CN204465555U CN 204465555 U CN204465555 U CN 204465555U CN 201520223978 U CN201520223978 U CN 201520223978U CN 204465555 U CN204465555 U CN 204465555U
- Authority
- CN
- China
- Prior art keywords
- password
- claimant
- calibration equipment
- user
- 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.)
- Expired - Fee Related
Links
Landscapes
- Telephonic Communication Services (AREA)
Abstract
The utility model relates to the voiceprint authentication apparatus based on time type dynamic password, comprise the password generated device, recording device, data sending device and the data sink that are arranged at client, password generated device is for generating password, recording device is used for the voice data that record request people reads password, and data sending device transmits it to server end; Server end comprises the user ID calibration equipment, password calibration equipment, the vocal print calibration equipment that connect successively, user ID calibration equipment is for judging the user ID of claimant, password calibration equipment is for judging the password that claimant reads, and vocal print calibration equipment is for judging whether claimant is registrant.Superior effect of the present utility model is: the dynamic password of generation does not need to transmit between server end and client with text or ciphertext, ensures that dynamic password is not intercepted and captured by the 3rd people in Internet Transmission; Binding time type dynamic password and voiceprint authentication technology, that has taken precautions against playback swarms into attack, and fail safe is high.
Description
Technical field
The utility model belongs to computer information technology field, relates to authenticating user identification technology, is specifically related to a kind of voiceprint authentication apparatus based on time type dynamic password.
Background technology
Along with maturation and the great development that is universal, smart mobile phone industry of mobile Internet, the service of mobile Internet has covered the many aspects of people's life.While providing convenient service to people, mobile Internet also brings the safety problem become increasingly conspicuous.
On the internet user identity is carried out to the method for certification at present, mainly comprise hardware digital certificate, E-token dynamic password card, short message verification code etc.Time type dynamic password is with user's landing time for enchancement factor, and conversion in an every 60 seconds dynamic password usually, in the process generating dynamic password, can not produce any communication, therefore password can not be intercepted in communication channel.Dynamic password is easy to use, is widely used in the fields such as enterprise, network game, finance.But dynamic password has only accomplished the certification to trusted terminal, certification can not be carried out to the identity of the people using these terminals.
Voiceprint authentication method uses sound groove recognition technology in e, utilizes the sound of people to carry out certification to user identity.Be compared to the authentication methods such as hardware digital certificate, E-token dynamic password card, short message verification code, voiceprint directly carries out certification to the identity of people, can solve the problem that subscriber equipment is stolen.
The same existing defects of existing voiceprint authentication technology, is difficult to the problem solving Replay Attack; If sound when someone uses the illegal means such as recording to get user authentication, uses this recording again to carry out certification.Voiceprint is difficult to take precautions against such attack.
Utility model content
The purpose of this utility model is to overcome deficiency of the prior art, provides a kind of voiceprint authentication apparatus based on time type dynamic password.
The utility model is achieved through the following technical solutions:
Based on the voiceprint authentication apparatus of time type dynamic password, comprise the password generated device being arranged at client, recording device, the data sending device be connected with described recording device and data sink, described password generated device is used for the password generating a random character sequence after claimant proposes authentication request, and described password is shown to claimant by display device, described recording device is used for the voice data that record request people reads above-mentioned password, described display device is connected with password generated device, described recording device by gather data transmission in network telephony to data sending device, described data sending device by network equipment by data transmission in network telephony to server end, described server end comprises the user ID calibration equipment connected successively, password calibration equipment, vocal print calibration equipment, whether described user ID calibration equipment meets authentication condition for the user ID judging claimant and input, described password calibration equipment for receive claimant voice data after, judge whether the password that claimant reads meets verification condition, described vocal print calibration equipment for receive claimant voice data after, the voice data of claimant and the sound-groove model of user ID are tested, judge that whether claimant is the registrant of the user ID of input, and check results is transferred to data sink and the display device of client by network equipment.
Described technical scheme is preferably, the built-in sound identification module of described password calibration equipment.
Described technical scheme is preferably, and the password generated step of described password generated device comprises: the first step calculates one-way hash function using the time of user ID and request authentication as input parameter, draws a random character sequence; Second step adds separator formative dynamics password in random character sequence, and is shown to claimant by display device, and described separator is space or horizontal line.
Described technical scheme is preferably, and the checking procedure of described user ID calibration equipment comprises: the first step inquires about described user ID whether in the ID storehouse of having recorded, and result, for being, carries out second step; Result is otherwise verifies unsuccessfully; Second step is inquired about described user ID and whether is less than at the authentication request number of times of current time the certification frequency threshold value that server end arranges, and result is for being then verification succeeds, and result is otherwise verifies unsuccessfully.
Described technical scheme is preferably, the checking procedure of described password calibration equipment comprises: described password calibration equipment adopts the password generated step identical with S01, using the time of user ID and request authentication as input parameter, calculate one-way hash function, generate a random character sequence, and adopt sound identification module that the voice data of claimant is converted to character string, the random character sequence of generation and the character string of conversion are compared, if two groups of character strings are identical, verification succeeds, otherwise then verify failure.
Described technical scheme is preferably, the checking procedure of described vocal print calibration equipment comprises: vocal print calibration equipment described in the first step extracts vocal print characteristic vector according to the voice data of claimant, comprise the short-term spectrum characteristic sequence extracting described voice data, for each the frame feature in above-mentioned characteristic sequence, calculate the posterior probability of each frame on global context model, adaptive training is carried out according to maximum posteriori criterion, obtain the gauss hybrid models of described request people, the Gaussian component of described gauss hybrid models is spliced into a high dimension vector, described high dimension vector is vocal print characteristic vector, second step calculates the inner product of the sound-groove model vector of described vocal print characteristic vector and user ID, and the inner product drawn is the vocal print similarity of claimant and request user ID, if vocal print similarity is greater than the threshold value pre-set, vocal print verification succeeds, on the contrary then vocal print verifies unsuccessfully.
Compared with prior art, superior effect of the present utility model is: the dynamic password that described password generated device generates does not need to transmit between server end and client with text or ciphertext, has ensured that described dynamic password is not intercepted and captured by the 3rd people in Internet Transmission; Binding time type dynamic password and voiceprint authentication technology simultaneously, authenticated user is when repeatedly carrying out voiceprint, and the content of its voice data is dynamic change, and that has taken precautions against playback swarms into attack, enhances the fail safe of certification.
Accompanying drawing explanation
Fig. 1 is the structural representation of the utility model based on the voiceprint authentication apparatus of time type dynamic password.
Accompanying drawing mark is as follows:
11-password generated device, 12-recording device, 13-data sending device, 14-data sink, 15-I/O device, 21-user ID calibration equipment, 22-password calibration equipment, 23-vocal print calibration equipment, 3-network equipment.
Embodiment
Below in conjunction with accompanying drawing, the utility model embodiment is described in further detail.
As shown in Figure 1, voiceprint authentication apparatus based on time type dynamic password described in the utility model, comprise the password generated device 11 being arranged at client, recording device 12, the data sending device 13 be connected with described recording device 12 and data sink 14, described password generated device 11 generates the password of a random character sequence after proposing authentication request for claimant, and described password is shown to claimant by display device (not shown), described display device is connected with password generated device 11, described recording device 12 reads the voice data of above-mentioned password for record request people, described recording device 12 by gather data transmission in network telephony to data sending device 13, described data sending device 13 passes through network equipment 3 by data transmission in network telephony to server end, described server end comprises the user ID calibration equipment 21, password calibration equipment 22, the vocal print calibration equipment 23 that connect successively, whether described user ID calibration equipment 21 meets authentication condition for the user ID judging claimant and input, described password calibration equipment 22 for receive claimant voice data after, judge whether the password that claimant reads meets verification condition, the built-in sound identification module (not shown) of described password calibration equipment 22, described vocal print calibration equipment 23 for receive claimant voice data after, the voice data of claimant and the sound-groove model of user ID are tested, judge that whether claimant is the registrant of the user ID of input, and check results is transferred to the data sink 14 of client by network equipment 3, show on the display device of client simultaneously.
I/O device 15 between described client and display device is prior art.
A kind of voiceprint authentication method based on time type dynamic password described in the utility model, have employed the above-mentioned voiceprint authentication apparatus based on time type dynamic password, and described authentication method comprises the following steps:
S01: described password generated device 11, after claimant proposes authentication request, generates the password of a random character sequence, and described password is shown to claimant by display device.
Before claimant proposes authentication request, need the sound-groove model of registered user; Comprise: according to user ID and the request time generation dynamic password of registered user; After user reads the voice data of password, extract the sound-groove model of vocal print characteristic vector as user of described voice data; User's typing one or many voice data when registering simultaneously.
The password generated step of described password generated device 11 comprises: the first step calculates one-way hash function using the time of user ID and request authentication as input parameter, draws a random character sequence; Second step adds separator formative dynamics password in random character sequence, and is shown to claimant by display device, and described separator is space or horizontal line.
When practical operation, password can be divided into many groups by adding of described separator, is convenient to claimant and reads.If dynamic password length is 8 characters, then this password is divided into 2 groups, often organizes 4 characters, in the middle of 2 groups, add 1 separator; If dynamic password length is 12 characters, then this password is divided into 3 groups, often organizes 4 characters, in each group, add 1 separator.
S02: described recording device 12 record request people reads the voice data of above-mentioned password, described recording device 12 is by the data transmission in network telephony of collection to data sending device 13, and described data sending device 13 passes through network equipment 3 by data transmission in network telephony to server end.
S03: whether the user ID of the user ID calibration equipment 21 check request people input of described server end meets authentication condition.
The checking procedure of described user ID calibration equipment 21 comprises: the first step inquires about described user ID whether in the ID storehouse of having recorded, and result, for being, carries out second step; Result is otherwise verifies unsuccessfully; Second step is inquired about described user ID and whether is less than at the authentication request number of times of current time the certification frequency threshold value that server end arranges, and result is for being then verification succeeds, and result is otherwise verifies unsuccessfully.
In actual design, counted 2 before request time minute, the request number of times sum in 2 minutes is as the authentication request number of times of this user ID in current time; Arranging of threshold value adjusts according to server end loading condition, as this threshold value is set to 5.
S04: described password calibration equipment 22 for receive claimant voice data after, whether the password that check request people reads meets verification condition.
The checking procedure of described password calibration equipment 22 comprises: described password calibration equipment 22 adopts the password generated step identical with S01, using the time of user ID and request authentication as input parameter, calculate one-way hash function, generate a random character sequence, and adopt sound identification module that the voice data of claimant is converted to character string, the random character sequence of generation and the character string of conversion are compared, if two groups of character strings are identical, verification succeeds, on the contrary then verify failure.
S05: described vocal print calibration equipment 23 for receive claimant voice data after, the voice data of claimant and the sound-groove model of user ID are verified, judge that whether claimant is the registrant of the user ID of input, and check results is transferred to the data sink 14 of client by network equipment 3, show on the display device of client simultaneously.
The checking procedure of described vocal print calibration equipment 23 comprises: vocal print calibration equipment 23 described in the first step extracts vocal print characteristic vector according to the voice data of claimant, comprise the short-term spectrum characteristic sequence extracting described voice data, for each the frame feature in above-mentioned characteristic sequence, calculate the posterior probability of each frame on global context model, adaptive training is carried out according to maximum posteriori criterion, obtain the gauss hybrid models of described request people, the Gaussian component of described gauss hybrid models is spliced into a high dimension vector, described high dimension vector is vocal print characteristic vector, second step calculates the inner product of the sound-groove model vector of described vocal print characteristic vector and user ID, and the inner product drawn is the vocal print similarity of claimant and request user ID, if vocal print similarity is greater than the threshold value pre-set, vocal print verification succeeds, on the contrary then vocal print verifies unsuccessfully.Described short-term spectrum feature adopts mel-frequency cepstrum coefficient or perception linear predictor coefficient.Described global context model is the dynamic password speech data of employing 1000 people, the gauss hybrid models obtained by EM Algorithm for Training.
The utility model is not limited to above-mentioned execution mode, and when not deviating from flesh and blood of the present utility model, any distortion that it may occur to persons skilled in the art that, improvement, replacement all fall into scope of the present utility model.
Claims (2)
1. based on the voiceprint authentication apparatus of time type dynamic password, it is characterized in that, comprise the password generated device being arranged at client, recording device, the data sending device be connected with described recording device and data sink, described password generated device is used for the password generating a random character sequence after claimant proposes authentication request, and described password is shown to claimant by display device, described recording device is used for the voice data that record request people reads above-mentioned password, described display device is connected with password generated device, described recording device by gather data transmission in network telephony to data sending device, described data sending device by network equipment by data transmission in network telephony to server end,
Described server end comprises the user ID calibration equipment connected successively, password calibration equipment, vocal print calibration equipment, whether described user ID calibration equipment meets authentication condition for the user ID judging claimant and input, described password calibration equipment for receive claimant voice data after, judge whether the password that claimant reads meets verification condition, described vocal print calibration equipment for receive claimant voice data after, the voice data of claimant and the sound-groove model of user ID are tested, judge that whether claimant is the registrant of the user ID of input, and check results is transferred to data sink and the display device of client by network equipment.
2. the voiceprint authentication apparatus based on time type dynamic password according to claim 1, is characterized in that, the built-in sound identification module of described password calibration equipment.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201520223978.4U CN204465555U (en) | 2015-04-14 | 2015-04-14 | Based on the voiceprint authentication apparatus of time type dynamic password |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201520223978.4U CN204465555U (en) | 2015-04-14 | 2015-04-14 | Based on the voiceprint authentication apparatus of time type dynamic password |
Publications (1)
Publication Number | Publication Date |
---|---|
CN204465555U true CN204465555U (en) | 2015-07-08 |
Family
ID=53672375
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201520223978.4U Expired - Fee Related CN204465555U (en) | 2015-04-14 | 2015-04-14 | Based on the voiceprint authentication apparatus of time type dynamic password |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN204465555U (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106448685A (en) * | 2016-10-09 | 2017-02-22 | 北京远鉴科技有限公司 | System and method for identifying voice prints based on phoneme information |
-
2015
- 2015-04-14 CN CN201520223978.4U patent/CN204465555U/en not_active Expired - Fee Related
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106448685A (en) * | 2016-10-09 | 2017-02-22 | 北京远鉴科技有限公司 | System and method for identifying voice prints based on phoneme information |
CN106448685B (en) * | 2016-10-09 | 2019-11-22 | 北京远鉴科技有限公司 | A kind of voiceprint authentication system and method based on phoneme information |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104834849B (en) | Dual-factor identity authentication method and system based on Application on Voiceprint Recognition and recognition of face | |
US9979721B2 (en) | Method, server, client and system for verifying verification codes | |
US10135818B2 (en) | User biological feature authentication method and system | |
CN104901808A (en) | Voiceprint authentication system and method based on time type dynamic password | |
US8862888B2 (en) | Systems and methods for three-factor authentication | |
US7970611B2 (en) | Speaker authentication in digital communication networks | |
CN107393541B (en) | Information verification method and device | |
US8234494B1 (en) | Speaker-verification digital signatures | |
CN106961418A (en) | Identity identifying method and identity authorization system | |
CN105577664B (en) | Password remapping method and system, client and server | |
US20170092276A1 (en) | Voiceprint Verification Method And Device | |
US20130262873A1 (en) | Method and system for authenticating remote users | |
CN103679452A (en) | Payment authentication method, device thereof and system thereof | |
CN105141615A (en) | Method and system for opening account remotely, authentication method and system | |
CN111883140A (en) | Authentication method, device, equipment and medium based on knowledge graph and voiceprint recognition | |
CN104092653B (en) | Data processing method and system | |
CN109802942A (en) | A kind of voiceprint authentication method and system, mobile terminal of secret protection | |
CN109005104B (en) | Instant messaging method, device, server and storage medium | |
CN104158664A (en) | Identity authentication method and system | |
CN103024706A (en) | Short message based device and short message based method for bidirectional multiple-factor dynamic identity authentication | |
CN108074575A (en) | Identity verification method and device based on recurrent neural network | |
CN106878324B (en) | Short message authentication method, short message authentication server and terminal | |
CN104821882A (en) | Network security verification method based on voice biometric features | |
US10936705B2 (en) | Authentication method, electronic device, and computer-readable program medium | |
US20120330663A1 (en) | Identity authentication system and method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
C56 | Change in the name or address of the patentee | ||
CP01 | Change in the name or title of a patent holder |
Address after: 100142 East 9A, Hui Hui building, No. 158 West Fourth Ring Road, Beijing, Haidian District Patentee after: Beijing Yuanjian Technologies Co.,Ltd. Address before: 100142 East 9A, Hui Hui building, No. 158 West Fourth Ring Road, Beijing, Haidian District Patentee before: TimeSafer (Beijing) Technology Co.,Ltd. |
|
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20150708 |
|
CF01 | Termination of patent right due to non-payment of annual fee |