CN104835497A - Voiceprint card swiping system and method based on dynamic password - Google Patents
Voiceprint card swiping system and method based on dynamic password Download PDFInfo
- Publication number
- CN104835497A CN104835497A CN201510175691.3A CN201510175691A CN104835497A CN 104835497 A CN104835497 A CN 104835497A CN 201510175691 A CN201510175691 A CN 201510175691A CN 104835497 A CN104835497 A CN 104835497A
- Authority
- CN
- China
- Prior art keywords
- unit
- password
- user
- voiceprint
- vocal print
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a voiceprint card swiping system and method based on a dynamic password. The card swiping system comprises a first password generating unit, a voice acquiring unit, a real-time transmitting unit, and a server end, wherein, the first password generating unit, the voice acquiring unit, and the real-time transmitting unit are disposed in a client end and the server end is composed of a real-time receiving unit, an online identification unit, a voiceprint registering unit, a password verifying unit, a voiceprint identification unit, a second password generating unit, and a card swiping result control unit. The voiceprint card swiping method comprises that a request is submitted; the first and the second password generating units generate passwords; a user reads voice data and transmits the voice data to the server end; the online identification unit extracts voice data information; the password verifying unit verifies the passwords; the voiceprint identification unit verifies a voiceprint characteristics and transmits a result to a display device. The voiceprint card swiping system and method have beneficial effects of by performing verification with the dynamic password and the voiceprint information, guaranteeing the safety and convenience of card swiping, reducing a possibility that a user password is stolen, cracked, and forgotten, having low requirements for hardware and bandwidth, and achieving good safety.
Description
Technical field
The present invention relates to gate inhibition and work attendance to check card technical field, be specifically related to a kind of vocal print punch card system based on dynamic password and method.
Background technology
Traditional punch card system is adopt following several mode: the forms such as numerical ciphers button, physics magnetic card are checked card, fingerprint is checked card substantially; Wherein, easily there is the problem that Cipher Strength is not high, password is forgotten, password is revealed in numerical ciphers key mode; Physics magnetic card is checked card and be there is possibility that is stolen or that lose, and security is not enough; Fingerprint is checked card and is not suitable for remote validation and checks card, and inquiry and management are checked card information inconvenience.
Sound groove recognition technology in e is the one of biological identification technology, and this technology by reflecting the speech parameter of speak human physiology and behavioural characteristic in speech waveform, and then identifies speaker ' s identity.
Sound groove recognition technology in e has two types in actual applications, and the first is that vocal print confirms, namely confirms when supposing identity whether voice to be identified belong to this supposition identity people, i.e. man-to-man identification, this is a kind of form that current internet, applications is the widest.The patent No. is that the Chinese patent of ZL201310123555.0 discloses and belongs to based on the identification confirmation system of dynamic password voice and method the form that vocal print confirms, but this identification confirmation system adopts server to generate dynamic password and is issued to the mode of client, is is easily intercepted and captured by the 3rd people in Internet Transmission; The second type is vocal print identification, does not namely know that sound to be identified belongs to whom in advance, needs the speaker ' s identity determining this sound from a large amount of sound-groove model, or judge that voice to be identified belong to outside collection, be i.e. the identification of one-to-many.
Be compared to fingerprint, the palm biological identification technology such as shape and iris, the advantage that the application of Application on Voiceprint Recognition has it special: the voice the first, containing vocal print feature obtain convenient, nature, and voiceprint extraction completes unconsciously, and therefore the acceptance level of user is higher; The second, the identification obtaining voice is with low cost, and use simple, a microphone can complete, when using communication apparatus more without the need to extra sound pick-up outfit; Three, be applicable to remote authentication, only need a microphone or mobile phone just by the identification of real-time performance remote identity; Four, compatible strong, as carried out content discriminating etc. by speech recognition, recognition accuracy and level of security can be improved.On the one hand, the single-factor authentication technology based on numerical password has been widely used in gate inhibition, work attendance product.On the other hand, the biometrics identification technology based on individual vocal print has more security and convenience than traditional punch card method, have anti-counterfeiting performance good, not easily to forge or stolen, with oneself " carrying " and use whenever and wherever possible, the advantages such as applicable remote identification.
In prior art, single-factor authentication technical security performance of checking card is low, can not satisfy the demands.
Summary of the invention
The object of the invention is to overcome deficiency of the prior art, a kind of vocal print punch card system based on dynamic password and method are provided, solve existing single factor test and to check card Problems existing in technology.
The present invention is achieved by the following technical solutions:
A kind of vocal print punch card system based on dynamic password, comprise and be arranged at client for generating the first password generation unit of dynamic password, the voice collecting unit for gathering speech data be connected with described first password generation unit, the real-time Transmission unit for real-time delivery of voice data be connected with described voice collecting unit, described real-time Transmission unit by voice data transmission to server end;
Described server end is provided with the real-time reception unit for real-time reception speech data, described real-time reception unit is connected with ONLINE RECOGNITION unit, described ONLINE RECOGNITION unit is connected with voiceprint registration unit, password verification unit, Application on Voiceprint Recognition unit respectively, described password verification unit is arranged between voiceprint registration unit and Application on Voiceprint Recognition unit, and be connected with described password verification unit, generate the second password generated unit of same challenge with the first password generation unit; Described voiceprint registration unit, Application on Voiceprint Recognition unit are connected with vocal print plate device and voiceprint registration unit, Application on Voiceprint Recognition unit are also connected with output control unit of checking card, described in output control unit of checking card be connected with the display device being arranged at client.
Described technical scheme is preferably, the built-in sound identification module of described password verification unit.
Described technical scheme is preferably, be provided with the speech compression unit for compressed voice data between described voice collecting unit and real-time Transmission unit, between described real-time reception unit and ONLINE RECOGNITION unit, be provided with the voice decompression units for decompressed speech data.Described speech compression unit and voice decompression units can be used for compression and the decompress(ion) of the speech data under various bandwidth condition.After described speech compression unit receives the speech data of described voice collecting unit, time span according to setting carries out compress speech, and by compression after speech data by network from client real-time Transmission, described voice decompression units carries out decompression processing to the voice signal received in real time.
Described first password generation unit, the second password generated unit all adopt identical dynamic password generating algorithm, by minute in units of generate random cipher string, the dynamic password namely generated; Described ONLINE RECOGNITION unit is used for sending the phonetic feature of speech data in line computation to client, provides vocal print modeling and carries out in line computation for the vocal print feature that Application on Voiceprint Recognition uses, improving user's registration and response speed of checking card; The dynamic password that described password verification unit is used for the first password generation unit, the second password generated unit generate verifies.
Described voiceprint registration unit is according to its voiceprint of the data under voice of user, and the dynamic password generated with the second password generated unit verifies, after the voiceprint of user succeeds in registration, by its voiceprint stored in vocal print plate device.The speech data of described Application on Voiceprint Recognition unit air exercise card user carries out identifying and confirming, the vocal print feature that described Application on Voiceprint Recognition unit calculates according to ONLINE RECOGNITION unit, and the sound-groove model of deriving in vocal print plate device carries out vocal print characteristic similarity coupling, obtain the user ID that similarity score is maximum, and the mark of this user ID is compared with the threshold value pre-set; If the mark of user ID is greater than threshold value, check card successfully, otherwise for checking card unsuccessfully.
Described output control unit of checking card derives the user ID of vocal print plate device in the process of user's voiceprint registration, carries out user ID repeatability and detects; After voiceprint registration success, user's voiceprint is imported vocal print plate device; Check card in test process user, the vocal print list and the vocal print template that derive user carry out vocal print Similarity Measure; Carry out voiceprint registration as user or check card in process, the object information of the success or not of record voiceprint registration, test success or not of checking card, be convenient to later stage searching and managing, and object information of being checked card by vocal print is sent to the display device being arranged at client.
A kind of vocal print punch card method based on dynamic password of the present invention, have employed the above-mentioned vocal print punch card system based on dynamic password, described punch card method at least comprises the following steps:
S01: user registers, submits the request of checking card to after registration;
S02: after the request of checking card of the second password generated cell response of server end, generate dynamic password, the first password generation unit of trigger clients generates identical dynamic password simultaneously, described dynamic password using minute in units of as initial conditions;
S03: user reads the dynamic password that the display device that is arranged at client provides, is gathered the speech data of user, is undertaken compressing and send it to server end by real-time Transmission unit by described speech compression unit by voice collecting unit;
S04: the real-time reception unit of described server end obtains the speech data of Real Time Compression and sends into online recognition unit after carrying out decompression processing to it, carries out On-line testing to the phonetic feature of the speech data of user; The phonetic feature of user is sent into Application on Voiceprint Recognition unit simultaneously, On-line testing is carried out to the vocal print feature of user;
S05: the phonetic feature comprising all speech frames that described ONLINE RECOGNITION unit exports, sound identification module in described password verification unit draws voice identification result according to the phonetic feature exported, and carries out consistency desired result with the dynamic password that the second password generated unit generates; If result is inconsistent, then the failure the object information of failure of checking card is fed back to client of checking card;
S06: described Application on Voiceprint Recognition unit extracts the vocal print feature of user and compares from the sound-groove model that vocal print plate device is derived, and obtains maximum vocal print similarity score and the user ID of correspondence thereof; If the mark of this user ID is greater than the decision threshold pre-set, check card successfully, otherwise for checking card unsuccessfully, feed back to client by described output control unit of checking card object information of checking card.
Described technical scheme is preferably, and in described S01, when user registers, at least comprises the following steps:
The first step: user submits registration request to by client, carries out user ID repeatability by server end and detects;
Second step: after the second password generated cell response registration request of server end, generates dynamic password at server end, and triggers the first password generation unit being arranged at client and generate identical password, minute being that unit is as initial conditions;
3rd step: user reads the dynamic password that the display device that is arranged at client provides, is gathered the speech data of user, is undertaken compressing and send it to server end by real-time Transmission unit by described speech compression unit by voice collecting unit;
4th step: the real-time reception unit of described server end obtains the speech data of Real Time Compression and sends into online recognition unit to after its decompression processing, carries out On-line testing to the phonetic feature of the speech data of user; The phonetic feature of user is sent into voiceprint registration unit simultaneously, On-line testing is carried out to the vocal print feature of user;
5th step: the phonetic feature comprising all speech frames that described ONLINE RECOGNITION unit exports, sound identification module in described password verification unit draws voice identification result according to the phonetic feature exported, and carries out consistency desired result with the dynamic password that the second password generated unit generates; If result is inconsistent, then registration failure the object information of registration failure is fed back to client;
6th step: described voiceprint registration unit extracts the vocal print feature of user, obtains complete vocal print feature, and sent into vocal print plate device establishment sound-groove model; If sound-groove model creates unsuccessfully, the object information of voiceprint registration failure is also fed back to client by voiceprint registration failure; If sound-groove model creates successfully, the voiceprint of user is joined in current vocal print plate device, equally the object information succeeded in registration is fed back to client.
Described technical scheme is preferably, and described phonetic feature is spectrum signature.
Described technical scheme is preferably, and described spectrum signature is mel-frequency cepstrum coefficient or perception linear predictor coefficient.
Compared with prior art, superior effect of the present invention is:
(1) carry out dual factors checking by the voiceprint of dynamic password and user, ensured the security and convenience of checking card, significantly reduce the possibility that user cipher is stolen, be cracked and forget;
(2) low to hardware requirement.Generate identical dynamic password at client and server end, and by minute in units of generate simultaneously, and the proof procedure of dynamic password is at server end, avoids dynamic password and issues in transmitting procedure and produce potential safety hazard; Server end, to the unified management of vocal print plate device, records user's registration of described punch card system and result of checking card, the inquiry of convenient air exercise card information and management;
(3) by client to input compress speech and real-time network transmission technology, reduce further the bandwidth requirement of Internet Transmission; Simultaneously at server end to the real-time decompress(ion) of data received and ONLINE RECOGNITION, drastically increase the response speed of user's registration and process of checking card;
(4) dynamic password of described first password generation unit and the second password generated unit stochastic generation does not transmit at client and server end, ensures that dynamic password is not intercepted and captured by the 3rd people in Internet Transmission.
Accompanying drawing explanation
Fig. 1 is the structural representation of the vocal print punch card system that the present invention is based on dynamic password;
Fig. 2 registers implementing procedure figure in the vocal print punch card method that the present invention is based on dynamic password;
Fig. 3 is the implementing procedure figure that checks card of the vocal print punch card method that the present invention is based on dynamic password;
Fig. 4 is the process flow diagram of Application on Voiceprint Recognition unit of the present invention identification and judgement.
Accompanying drawing mark is as follows:
11-first password generation unit, 12-voice collecting unit, 13-real-time Transmission unit, 14-display device, 15-speech compression unit, 21-real-time reception unit, 22-ONLINE RECOGNITION unit, 23-voiceprint registration unit, 24-password verification unit, 25-Application on Voiceprint Recognition unit, 26-second password generated unit, 27-vocal print plate device, 28-check card output control unit, 29-voice decompression units.
Embodiment
Below in conjunction with accompanying drawing, the specific embodiment of the invention is described in further detail.
As shown in Figure 1, a kind of vocal print punch card system based on dynamic password of the present invention, comprise and be arranged at client for generating the first password generation unit 11 of dynamic password, the voice collecting unit 12 for gathering speech data be connected with described first password generation unit 11, the real-time Transmission unit 13 for real-time delivery of voice data be connected with described voice collecting unit 12, described real-time Transmission unit 13 by voice data transmission to server end.
Described server end is provided with the real-time reception unit 21 for real-time reception speech data, described real-time reception unit 21 is connected 22 with ONLINE RECOGNITION unit, described ONLINE RECOGNITION unit 22 is connected with voiceprint registration unit 23, password verification unit 24, Application on Voiceprint Recognition unit 25 respectively, described password verification unit 24 is arranged between voiceprint registration unit 23 and Application on Voiceprint Recognition unit 25, and be connected with described password verification unit 24, generate the second password generated unit 26 of same challenge with the first password generation unit 11; Described voiceprint registration unit 23, Application on Voiceprint Recognition unit 25 are connected with vocal print plate device 27 and voiceprint registration unit 23, Application on Voiceprint Recognition unit 25 are also connected with output control unit 28 of checking card, described in output control unit 28 of checking card be connected with the display device 14 being arranged at client.The built-in sound identification module (not shown) of described password verification unit 24 is prior art.Be provided with the speech compression unit 15 for compressed voice data between described voice collecting unit 12 and real-time Transmission unit 13, between described real-time reception unit 21 and ONLINE RECOGNITION unit 22, be provided with the voice decompression units 29 for decompressed speech data.
When described voiceprint registration unit 23 is registered, by the request of server end customer in response end, described second password generated unit 26 and the first password generation unit 11 by minute in units of generate the dynamic password of current time, user reads the dynamic password that the first password generation unit 11 generates, described voice collecting unit 12 gathers the speech data being read above-mentioned dynamic password by user, and is sent to server end; Described server end carries out the verification of dynamic password text coherence to the voice messaging of user, verify by after to create the sound-groove model of active user at vocal print plate device 27, and recording user registration daily record, the object information succeeded in registration is fed back to client simultaneously.
During the vocal print feature identification of described Application on Voiceprint Recognition unit 25 couples of users, by the request of server end customer in response end, described second password generated unit 26 and the first password generation unit 11 by minute in units of generate the dynamic password of current time, user reads the dynamic password that the first password generation unit 11 generates, described voice collecting unit 12 gathers the speech data being read above-mentioned dynamic password by user, and is sent to server end; Described server end carries out the verification of dynamic password text coherence to the voice messaging of user, described dynamic password verification is by the sound-groove model of all users and the user ID information of correspondence in rear derivation current vocal print plate device 27, carry out vocal print Similarity Measure, and obtain the maximum user ID of vocal print characteristic similarity, its vocal print similarity score and the threshold value pre-set are compared, if be greater than threshold value, active user checks card successfully, and recording user is checked card daily record; Otherwise check card unsuccessfully, and will to check card successfully or the object information of failure feeds back to the display device 14 of client.
A kind of vocal print punch card method based on dynamic password of the present invention, needed to carry out user's registration, as shown in Figure 2, comprises the following steps before carrying out punching operation:
The first step: user submits registration request to by client, carries out user ID repeatability by server end and detects; If comprise the user name of registration request in the vocal print plate device 27 of server end, then user is pointed out to re-register.
Second step: after the second password generated unit 26 of server end responds registration request, generates dynamic password at server end, and triggers the first password generation unit 11 being arranged at client and generate identical password, minute being that unit is as initial conditions.The generation of described dynamic password adopts existing known mature technology to realize, and it realizes principle and means belong to technology well-known to those skilled in the art.
3rd step: user reads the dynamic password that the display device 14 that is arranged at client provides, the speech data of user is gathered by voice collecting unit 12, by setting-up time length, speech data is compressed, until all speech data compressions are complete, undertaken compressing by described speech compression unit 15 and sent it to server end by real-time Transmission unit 13.
When specifically setting, the time span of above-mentioned setting is 0.5 second; Compress speech adopts Speex mode, and this is a set of for the increasing income of voice, audio compression format without patent protection.Speex provides a high performance encoding and decoding speech method, based on CELP (Code Excited Linear Prediction, Qualcomm Code Excited Linear Prediction (QCELP)) and specially for code check designs in the compress speech of 2kbps to 44kbps, be particluarly suitable for network application, vocal print punch card system provided by the invention can be run under the network environment of low-bandwidth condition.
4th step: the real-time reception unit 21 of described server end obtains the speech data of Real Time Compression and to sending into online recognition unit 22 after its decompression processing, carries out On-line testing to the phonetic feature of the speech data of user; The phonetic feature of user is sent into voiceprint registration unit 23 simultaneously, On-line testing is carried out to the vocal print feature of user.
In actual design, described phonetic feature adopts spectrum signature, as mel-frequency cepstrum coefficient MFCC, perception linear prediction PLP.Adopt cepstral mean normalized square mean (CMVN) to carry out in the process of channel abatement, in line computation, the present invention adopts incremental mode to calculate its average and variance parameter, and detailed process is:
For given one section of voice, comprise N (N>1) frame phonetic feature x
1 ..., N, its average μ
nand variance
be respectively:
Assuming that the voice that online process one section is new, it comprises M-N (M-N>=1) frame phonetic feature x
n+1 ..., M, the μ of history voice can be utilized
nwith
parameter, and new M-N frame phonetic feature, by the average μ that incremental mode estimation current time is corresponding
mand variance parameter
and no longer need to carry out double counting to historical data:
Wherein, x refers to phonetic feature, x
1 ..., N, x
n+1 ..., Min 1 ..., N, N+1 ..., M is the sequence number of phonetic feature frame number.
In actual design, after obtaining the spectrum signature (MFCC or PLP) of Application on Voiceprint Recognition, the present invention adopts total difference spatial model (Total variability model, TVM, also referred to as iVector) to carry out vocal print modeling and identification.Higher-dimension super vector by speaker and passage integrally, assuming that this entirety distributes in a low n-dimensional subspace n, is dropped to iVector dimension by total difference spatial model, is generally 300-800 dimension.Then in lower dimensional space, linear discriminant analysis (Linear discriminant analysis is passed through, LDA) and covariance normalization in class (Within-class covariancenormalization, WCCN) be separated speaker and channel information.
In total difference spatial model leaching process, need 0 rank and the 1 rank statistic of extracting BW (Baum-Welch).Common BW statistic extracting method is based on global context model (UBM), calculates the posterior probability of every frame phonetic feature on UBM model, obtains 0 rank and 1 rank statistic accordingly.Embodiment of the present invention adopts deep neural network method to extract BW statistic, effectively extracts the information of speaker.Deep neural network simulates human brain to the process of sound signal processing by the network structure of multilayer, more abstract high-rise representation feature is formed by combination low layer spectrum signature, can better high-level speaker information with the anti-interference learning to comprise in voice signal.
5th step: the phonetic feature comprising all speech frames that described ONLINE RECOGNITION unit 22 exports.By adopting Viterbi (Viterbi) decoding algorithm, sound identification module in described password verification unit 24 draws voice identification result according to the phonetic feature exported, and carries out consistency desired result with the dynamic password that the second password generated unit 26 generates; If result is inconsistent, then registration failure the object information of registration failure is fed back to client.
6th step: described voiceprint registration unit 23 extracts the vocal print feature of user, obtains complete vocal print feature, and is sent into vocal print plate device 27 and create sound-groove model; If sound-groove model creates unsuccessfully, the object information of voiceprint registration failure is also fed back to client by voiceprint registration failure; If sound-groove model creates successfully, the voiceprint of user is joined in current vocal print plate device 27, equally the object information succeeded in registration is fed back to client.
User, after completing above-mentioned registration step, just can carry out punching operation, as shown in Figure 3, specifically comprise the following steps:
S01: user submits the request of checking card to.
S02: generate dynamic password after the second password generated unit 26 response request of checking card of server end, simultaneously the first password generation unit 11 of trigger clients generates identical dynamic password, described dynamic password using minute in units of as initial conditions.
S03: user reads the dynamic password that the display device 14 that is arranged at client provides, and is gathered the speech data of user by voice collecting unit 12.Undertaken compressing by described speech compression unit 15 and sent it to server end by real-time Transmission unit 13; Speex voice coding/decoding algorithms is adopted to carry out real-time voice compression during concrete enforcement.
S04: the real-time reception unit 21 of described server end obtains the speech data of Real Time Compression and sends into online recognition unit 22 after carrying out decompression processing to it, carries out On-line testing to the phonetic feature of the speech data of user; The phonetic feature of user is sent into Application on Voiceprint Recognition unit 25 simultaneously, On-line testing is carried out to the vocal print feature of user; The account form of above-mentioned ONLINE RECOGNITION unit 22 is identical with registration process.
S05: the phonetic feature comprising all speech frames that described ONLINE RECOGNITION unit 22 exports, sound identification module in described password verification unit 24 draws voice identification result according to the phonetic feature exported, and carries out consistency desired result with the dynamic password that the second password generated unit 26 generates; If result is inconsistent, then the failure the object information of failure of checking card is fed back to client of checking card.
S06: described Application on Voiceprint Recognition unit 25 extracts the vocal print feature of user and compares from the sound-groove model that vocal print plate device 27 is derived, and obtains maximum vocal print similarity score and the user ID of correspondence thereof; If the mark of this user ID is greater than the decision threshold pre-set, check card successfully, otherwise for checking card unsuccessfully, feed back to client by described output control unit of checking card object information of checking card, specifically refer to the process flow diagram of Application on Voiceprint Recognition unit identification as shown in Figure 4 and judgement.
Vocal print punch card system based on dynamic password of the present invention and method, be applicable to gate inhibition and attendance checking system, and verified by the dual factors of dynamic password and individual vocal print, the identity registration and the vocal print that realize user are checked card.The present invention adopts the technical scheme of compress speech and decompress(ion), ensure that the connection speed of external network; Adopt ONLINE RECOGNITION unit to carry out speech recognition and Application on Voiceprint Recognition, improve user's registration and user and to check card the response speed of testing; Meanwhile, the present invention adopts the dual factors checking based on dynamic password and the individual vocal print feature of user also to contribute to improving security.
The present invention is not limited to above-mentioned embodiment, and when not deviating from flesh and blood of the present invention, any distortion that it may occur to persons skilled in the art that, improvement, replacement all fall into scope of the present invention.
Claims (7)
1. the vocal print punch card system based on dynamic password, it is characterized in that, comprise and be arranged at client for generating the first password generation unit of dynamic password, the voice collecting unit for gathering speech data be connected with described first password generation unit, the real-time Transmission unit for real-time delivery of voice data be connected with described voice collecting unit, described real-time Transmission unit by voice data transmission to server end;
Described server end is provided with the real-time reception unit for real-time reception speech data, described real-time reception unit is connected with ONLINE RECOGNITION unit, described ONLINE RECOGNITION unit is connected with voiceprint registration unit, password verification unit, Application on Voiceprint Recognition unit respectively, described password verification unit is arranged between voiceprint registration unit and Application on Voiceprint Recognition unit, and be connected with described password verification unit, generate the second password generated unit of same challenge with the first password generation unit; Described voiceprint registration unit, Application on Voiceprint Recognition unit are connected with vocal print plate device and voiceprint registration unit, Application on Voiceprint Recognition unit are also connected with output control unit of checking card, described in output control unit of checking card be connected with the display device being arranged at client.
2. a kind of vocal print punch card system based on dynamic password according to claim 1, is characterized in that, the built-in sound identification module of described password verification unit.
3. a kind of vocal print punch card system based on dynamic password according to claim 1, it is characterized in that, be provided with the speech compression unit for compressed voice data between described voice collecting unit and real-time Transmission unit, between described real-time reception unit and ONLINE RECOGNITION unit, be provided with the voice decompression units for decompressed speech data.
4., based on a vocal print punch card method for dynamic password, adopt the arbitrary described vocal print punch card system based on dynamic password of claim 1-3, it is characterized in that, described punch card method at least comprises the following steps:
S01: user registers, submits the request of checking card to after registration;
S02: after the request of checking card of the second password generated cell response of server end, generate dynamic password, the first password generation unit of trigger clients generates identical dynamic password simultaneously, described dynamic password using minute in units of as initial conditions;
S03: user reads the dynamic password that the display device that is arranged at client provides, is gathered the speech data of user, is undertaken compressing and send it to server end by real-time Transmission unit by described speech compression unit by voice collecting unit;
S04: the real-time reception unit of described server end obtains the speech data of Real Time Compression and sends into online recognition unit after carrying out decompression processing to it, carries out On-line testing to the phonetic feature of the speech data of user; The phonetic feature of user is sent into Application on Voiceprint Recognition unit simultaneously, On-line testing is carried out to the vocal print feature of user;
S05: the phonetic feature comprising all speech frames that described ONLINE RECOGNITION unit exports, sound identification module in described password verification unit draws voice identification result according to the phonetic feature exported, and carries out consistency desired result with the dynamic password that the second password generated unit generates; If result is inconsistent, then the failure the object information of failure of checking card is fed back to client of checking card;
S06: described Application on Voiceprint Recognition unit extracts the vocal print feature of user and compares from the sound-groove model that vocal print plate device is derived, and obtains maximum vocal print similarity score and the user ID of correspondence thereof; If the mark of this user ID is greater than the decision threshold pre-set, check card successfully, otherwise for checking card unsuccessfully, feed back to client by described output control unit of checking card object information of checking card.
5. a kind of vocal print punch card method based on dynamic password according to claim 4, is characterized in that,
In described S01, when user registers, at least comprise the following steps:
The first step: user submits registration request to by client, carries out user ID repeatability by server end and detects;
Second step: after the second password generated cell response registration request of server end, generates dynamic password at server end, and triggers the first password generation unit being arranged at client and generate identical password, minute being that unit is as initial conditions;
3rd step: user reads the dynamic password that the display device that is arranged at client provides, is gathered the speech data of user, is undertaken compressing and send it to server end by real-time Transmission unit by described speech compression unit by voice collecting unit;
4th step: the real-time reception unit of described server end obtains the speech data of Real Time Compression and sends into online recognition unit to after its decompression processing, carries out On-line testing to the phonetic feature of the speech data of user; The phonetic feature of user is sent into voiceprint registration unit simultaneously, On-line testing is carried out to the vocal print feature of user;
5th step: the phonetic feature comprising all speech frames that described ONLINE RECOGNITION unit exports, sound identification module in described password verification unit draws voice identification result according to the phonetic feature exported, and carries out consistency desired result with the dynamic password that the second password generated unit generates; If result is inconsistent, then registration failure the object information of registration failure is fed back to client;
6th step: described voiceprint registration unit extracts the vocal print feature of user, obtains complete vocal print feature, and sent into vocal print plate device establishment sound-groove model; If sound-groove model creates unsuccessfully, the object information of voiceprint registration failure is also fed back to client by voiceprint registration failure; If sound-groove model creates successfully, the voiceprint of user is joined in current vocal print plate device, equally the object information succeeded in registration is fed back to client.
6. a kind of vocal print punch card method based on dynamic password according to claim 4 or 5, it is characterized in that, described phonetic feature is spectrum signature.
7. a kind of vocal print punch card method based on dynamic password according to claim 6, is characterized in that, described spectrum signature is mel-frequency cepstrum coefficient or perception linear predictor coefficient.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510175691.3A CN104835497A (en) | 2015-04-14 | 2015-04-14 | Voiceprint card swiping system and method based on dynamic password |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510175691.3A CN104835497A (en) | 2015-04-14 | 2015-04-14 | Voiceprint card swiping system and method based on dynamic password |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104835497A true CN104835497A (en) | 2015-08-12 |
Family
ID=53813337
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510175691.3A Pending CN104835497A (en) | 2015-04-14 | 2015-04-14 | Voiceprint card swiping system and method based on dynamic password |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104835497A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105575394A (en) * | 2016-01-04 | 2016-05-11 | 北京时代瑞朗科技有限公司 | Voiceprint identification method based on global change space and deep learning hybrid modeling |
| CN105719370A (en) * | 2016-01-18 | 2016-06-29 | 上海交通大学 | Voiceprint verification system and method for parking lot |
| CN106251874A (en) * | 2016-07-27 | 2016-12-21 | 深圳市鹰硕音频科技有限公司 | A kind of voice gate inhibition and quiet environment monitoring method and system |
| CN106683661A (en) * | 2015-11-05 | 2017-05-17 | 阿里巴巴集团控股有限公司 | Role separation method and device based on voice |
| CN109801393A (en) * | 2018-12-27 | 2019-05-24 | 广州师盛展览有限公司 | A kind of artificial intelligence exhibition based on voice recognition is registered system |
| CN109872428A (en) * | 2019-02-25 | 2019-06-11 | 深圳市奇脉电子技术有限公司 | A kind of high degree of safety intelligent door lock and unlocking mode |
| CN111091638A (en) * | 2019-11-25 | 2020-05-01 | 星络智能科技有限公司 | Storage medium, intelligent door lock and authentication method thereof |
| CN111311774A (en) * | 2019-12-10 | 2020-06-19 | 上海明略人工智能(集团)有限公司 | Sign-in method and system based on voice recognition |
| CN113793425A (en) * | 2020-12-31 | 2021-12-14 | 北京京东尚科信息技术有限公司 | Patrol checking-in method and related equipment |
| CN115037710A (en) * | 2022-06-06 | 2022-09-09 | 中国工商银行股份有限公司 | Message transmission method and device |
| CN116071842A (en) * | 2022-11-21 | 2023-05-05 | 深聪半导体(江苏)有限公司 | Voiceprint-based card punching method and system |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101420297A (en) * | 2008-09-08 | 2009-04-29 | 北京飞天诚信科技有限公司 | Method and system for negotiating cipher |
| CN101459516A (en) * | 2009-02-20 | 2009-06-17 | 浙江工业大学 | Dynamic password safe login method |
| CN101800644A (en) * | 2010-01-11 | 2010-08-11 | 上海众烁信息科技有限公司 | Computer security protection system and method based on dynamic countersign |
| CN103020098A (en) * | 2012-07-11 | 2013-04-03 | 腾讯科技（深圳）有限公司 | Navigation service searching method with speech recognition function |
| CN103327158A (en) * | 2012-03-19 | 2013-09-25 | 上海博路信息技术有限公司 | Voice recognition locking and unlocking method |
| CN103986725A (en) * | 2014-05-29 | 2014-08-13 | 中国农业银行股份有限公司 | Client side, server side and identity authentication system and method |
| CN104158664A (en) * | 2014-08-11 | 2014-11-19 | 北京唐桓科技发展有限公司 | Identity authentication method and system |
| CN204494911U (en) * | 2014-12-31 | 2015-07-22 | 广东欧科空调制冷有限公司 | The oil-balancing system of Multi-compressor parallel unit |
-
2015
- 2015-04-14 CN CN201510175691.3A patent/CN104835497A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101420297A (en) * | 2008-09-08 | 2009-04-29 | 北京飞天诚信科技有限公司 | Method and system for negotiating cipher |
| CN101459516A (en) * | 2009-02-20 | 2009-06-17 | 浙江工业大学 | Dynamic password safe login method |
| CN101800644A (en) * | 2010-01-11 | 2010-08-11 | 上海众烁信息科技有限公司 | Computer security protection system and method based on dynamic countersign |
| CN103327158A (en) * | 2012-03-19 | 2013-09-25 | 上海博路信息技术有限公司 | Voice recognition locking and unlocking method |
| CN103020098A (en) * | 2012-07-11 | 2013-04-03 | 腾讯科技（深圳）有限公司 | Navigation service searching method with speech recognition function |
| CN103986725A (en) * | 2014-05-29 | 2014-08-13 | 中国农业银行股份有限公司 | Client side, server side and identity authentication system and method |
| CN104158664A (en) * | 2014-08-11 | 2014-11-19 | 北京唐桓科技发展有限公司 | Identity authentication method and system |
| CN204494911U (en) * | 2014-12-31 | 2015-07-22 | 广东欧科空调制冷有限公司 | The oil-balancing system of Multi-compressor parallel unit |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106683661B (en) * | 2015-11-05 | 2021-02-05 | 阿里巴巴集团控股有限公司 | Role separation method and device based on voice |
| CN106683661A (en) * | 2015-11-05 | 2017-05-17 | 阿里巴巴集团控股有限公司 | Role separation method and device based on voice |
| CN105575394A (en) * | 2016-01-04 | 2016-05-11 | 北京时代瑞朗科技有限公司 | Voiceprint identification method based on global change space and deep learning hybrid modeling |
| CN105719370A (en) * | 2016-01-18 | 2016-06-29 | 上海交通大学 | Voiceprint verification system and method for parking lot |
| CN105719370B (en) * | 2016-01-18 | 2018-06-12 | 上海交通大学 | Parking lot voiceprint verification system and its method |
| CN106251874A (en) * | 2016-07-27 | 2016-12-21 | 深圳市鹰硕音频科技有限公司 | A kind of voice gate inhibition and quiet environment monitoring method and system |
| WO2018018906A1 (en) * | 2016-07-27 | 2018-02-01 | 深圳市鹰硕音频科技有限公司 | Voice access control and quiet environment monitoring method and system |
| CN109801393A (en) * | 2018-12-27 | 2019-05-24 | 广州师盛展览有限公司 | A kind of artificial intelligence exhibition based on voice recognition is registered system |
| CN109872428A (en) * | 2019-02-25 | 2019-06-11 | 深圳市奇脉电子技术有限公司 | A kind of high degree of safety intelligent door lock and unlocking mode |
| CN111091638A (en) * | 2019-11-25 | 2020-05-01 | 星络智能科技有限公司 | Storage medium, intelligent door lock and authentication method thereof |
| CN111311774A (en) * | 2019-12-10 | 2020-06-19 | 上海明略人工智能(集团)有限公司 | Sign-in method and system based on voice recognition |
| CN113793425A (en) * | 2020-12-31 | 2021-12-14 | 北京京东尚科信息技术有限公司 | Patrol checking-in method and related equipment |
| CN115037710A (en) * | 2022-06-06 | 2022-09-09 | 中国工商银行股份有限公司 | Message transmission method and device |
| CN116071842A (en) * | 2022-11-21 | 2023-05-05 | 深聪半导体(江苏)有限公司 | Voiceprint-based card punching method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104835497A (en) | Voiceprint card swiping system and method based on dynamic password | |
| CN104834849B (en) | Dual-factor identity authentication method and system based on Application on Voiceprint Recognition and recognition of face | |
| WO2018166187A1 (en) | Server, identity verification method and system, and a computer-readable storage medium | |
| WO2018113526A1 (en) | Face recognition and voiceprint recognition-based interactive authentication system and method | |
| CN106847292B (en) | Method for recognizing sound-groove and device | |
| WO2017197953A1 (en) | Voiceprint-based identity recognition method and device | |
| Bailador et al. | Analysis of pattern recognition techniques for in-air signature biometrics | |
| CN110750774B (en) | Identity recognition method and device | |
| CN101256700A (en) | ATM of mixed type user identification authentication | |
| JP4937661B2 (en) | Mobile personal authentication method and electronic commerce method | |
| CN108074310A (en) | Voice interactive method and intelligent lock administration system based on sound identification module | |
| CN103106717A (en) | Intelligent warehouse voice control doorkeeper system based on voiceprint recognition and identity authentication method thereof | |
| CN102708867A (en) | Method and system for identifying faked identity by preventing faked recordings based on voiceprint and voice | |
| CN107886958A (en) | Express cabinet pickup method and device based on voiceprint | |
| CN113823293B (en) | Speaker recognition method and system based on voice enhancement | |
| CN110866234B (en) | Identity verification system based on multiple biological characteristics | |
| CN106991312A (en) | Internet based on Application on Voiceprint Recognition is counter to cheat authentication method | |
| CN204576520U (en) | Based on the Dual-factor identity authentication device of Application on Voiceprint Recognition and recognition of face | |
| CN204496911U (en) | A kind of vocal print punch-card device based on dynamic password | |
| CN201955775U (en) | USB identity authentication system based on voiceprint identification | |
| US20220398590A1 (en) | Payment verification method and payment verification system | |
| Jian et al. | An embedded voiceprint recognition system based on GMM | |
| CN209471456U (en) | A kind of Dual-factor identity authentication system of Application on Voiceprint Recognition and recognition of face | |
| CN112967724A (en) | Long-sequence biological Hash authentication method based on feature fusion | |
| CN111881432A (en) | Bank identity recognition system based on voice recognition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 100142 East 9A, Hui Hui building, No. 158 West Fourth Ring Road, Beijing, Haidian District Applicant after: Beijing Yuanjian Technologies Co.,Ltd. Address before: 100142 East 9A, Hui Hui building, No. 158 West Fourth Ring Road, Beijing, Haidian District Applicant before: Shidai Yibao (Beijing) Technology Co., Ltd. |
|
| COR | Change of bibliographic data | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150812 |