CN114003885B - Intelligent voice authentication method, system and storage medium - Google Patents

Intelligent voice authentication method, system and storage medium Download PDF

Info

Publication number
CN114003885B
CN114003885B CN202111284667.5A CN202111284667A CN114003885B CN 114003885 B CN114003885 B CN 114003885B CN 202111284667 A CN202111284667 A CN 202111284667A CN 114003885 B CN114003885 B CN 114003885B
Authority
CN
China
Prior art keywords
signal
voice
chest
user
thoracic cavity
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.)
Active
Application number
CN202111284667.5A
Other languages
Chinese (zh)
Other versions
CN114003885A (en
Inventor
陈艳姣
徐文渊
薛梦
孟紫云
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang University ZJU
Original Assignee
Zhejiang University ZJU
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Zhejiang University ZJU filed Critical Zhejiang University ZJU
Priority to CN202111284667.5A priority Critical patent/CN114003885B/en
Publication of CN114003885A publication Critical patent/CN114003885A/en
Application granted granted Critical
Publication of CN114003885B publication Critical patent/CN114003885B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
    • G06F21/31—User authentication
    • G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
    • A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/048—Activation functions
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/02—Feature extraction for speech recognition; Selection of recognition unit
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/08—Speech classification or search
    • G10L15/16—Speech classification or search using artificial neural networks
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
    • G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
    • G10L25/03—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters
    • G10L25/24—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters the extracted parameters being the cepstrum
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
    • G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
    • G10L25/27—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique
    • G10L25/30—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique using neural networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Acoustics & Sound (AREA)
  • Evolutionary Computation (AREA)
  • Multimedia (AREA)
  • Biophysics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Mathematical Physics (AREA)
  • Computing Systems (AREA)
  • Computer Security & Cryptography (AREA)
  • Data Mining & Analysis (AREA)
  • Signal Processing (AREA)
  • Physiology (AREA)
  • Animal Behavior & Ethology (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Surgery (AREA)
  • Medical Informatics (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Pathology (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Dentistry (AREA)
  • Computer Hardware Design (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

The invention discloses an intelligent voice authentication method, an intelligent voice authentication system and a storage medium, wherein the method comprises the following steps: s1: collecting a mixed signal sent by a user; s2: extracting the mixed signal to obtain a corresponding voice signal and a corresponding chest cavity action signal; s3: extracting the characteristics of the voice signal and the thoracic cavity action signal obtained in the step S2; s4: performing verification by using the characteristics obtained in the step S2; the invention relates the detected chest movement with the voice signal for identity authentication.

Description

Intelligent voice authentication method, system and storage medium
Technical Field
The invention relates to the technical field of intelligent authentication, in particular to an intelligent voice authentication method, an intelligent voice authentication system and a storage medium.
Background
Currently, voice-based authentication is ubiquitous on smart devices, where a voice assistant uses a wake-up word to activate the device and authenticate the user. With the help of a voice assistant, a user can perform various operations including sending messages, making phone calls, and viewing emails, etc.
However, in order to make the authentication process easier, various methods for performing voice-based authentication using only a smart phone without adding an additional device have been proposed, but voice-based authentication is vulnerable to replay attacks.
Therefore, how to provide an intelligent voice authentication method capable of solving the above problems is a problem that needs to be solved urgently by those skilled in the art.
Disclosure of Invention
In view of the above, the present invention provides an intelligent voice authentication method, system and storage medium, which associate the detected chest movement with a voice signal for identity authentication, have a long working distance, do not require any additional sensor, are very practical, have a high accuracy in recognition under various environments, and can well resist replay attack and imitation attack.
In order to achieve the purpose, the invention adopts the following technical scheme:
an intelligent voice authentication method, comprising:
s1: collecting a mixed signal sent by a user;
s2: extracting the mixed signal to obtain a corresponding voice signal and a corresponding chest cavity action signal;
s3: extracting the characteristics of the voice signal and the thoracic action signal obtained in the step S2;
s4: and performing verification by using the characteristics obtained in the step S3.
Preferably, the step S2 specifically includes: and processing the mixed signal by using a high-pass filter to obtain a corresponding voice signal, and processing the mixed signal by using a low-pass filter to obtain a corresponding thoracic action signal.
Preferably, the step S3 specifically includes:
s31: establishing a background model to process the voice signal, and smoothing the voice signal processed by the background model by using a sliding window to obtain a final voice command and a voice command interval;
s32: and establishing an adaptive filter to separate and reduce noise of the chest motion signal to obtain a final chest motion signal.
Preferably, the step S4 includes:
s41: identifying the thoracic cavity motion signal processed in the step S32, determining whether a user is present, and if so, proceeding to step S42;
s42: segmenting the chest cavity action signal according to the interval of the voice command, and acquiring a correlation coefficient of the chest cavity action signal and the voice command;
s43: and extracting features from the correlation coefficient, and training a deep neural network by using the features for the identification of the user.
Further, the present invention also provides an intelligent voice authentication system, comprising:
the data acquisition module is used for acquiring voice signals and thoracic cavity action signals of a user;
the data analysis module is connected with the data acquisition module and is used for analyzing and processing the voice signal and the thoracic cavity action signal;
the characteristic extraction module is connected with the data analysis module and is used for extracting the characteristics of the voice signal and the thoracic cavity action signal which are analyzed and processed;
and the intelligent terminal is connected with the feature extraction module and used for verifying according to the extracted features.
Preferably, the data analysis module includes:
a first data processing unit for processing the voice signal;
a second data processing unit for processing the chest action signal.
Preferably, the feature extraction module includes:
the voice signal extraction unit is used for extracting the processed voice signal to obtain a voice instruction;
and the thoracic cavity action signal extraction unit is used for extracting the processed voice signal to obtain a thoracic cavity action signal.
Further, the present invention also provides a computer readable storage medium having stored thereon computer instructions which, when executed, implement the method of any one of the above.
Compared with the prior art, the invention provides the intelligent voice authentication method, the intelligent voice authentication system and the storage medium, the detected chest movement is associated with the voice signal for identity authentication, the intelligent voice authentication method and the intelligent voice authentication system have longer working distance, do not need to install any additional sensor, are very practical, have higher accuracy rate of correct identification in various environments, and can better resist replay attack and imitation attack.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a flow chart of an intelligent voice authentication method according to the present invention;
fig. 2 is a schematic block diagram of a structure of an intelligent voice authentication device according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, an embodiment of the present invention discloses an intelligent voice authentication method, which can be implemented by an intelligent terminal, and specifically includes:
s1: collecting mixed signals sent by a user, and sending a pilot sound signal S (t) A by using an intelligent terminal 0 sin (2 π ft + φ) (where, A 0 Is amplitude, f is frequency, and phi is initial phase), the signal received by the smart terminal is a mixed signal after passing through the chest of the user and other reflections of objects, which can be expressed as:
Figure BDA0003332407650000041
in the formula, omega I Is the set of all reflected acoustic signal paths, A i f i And
Figure BDA0003332407650000042
respectively the amplitude, frequency and phase of path i.
The mixed signal mainly comprises three parts, namely a chest reflected signal, a body reflected signal and an environment object reflected signal, all frequency shifts of the pilot signal under the Doppler effect are considered, and a specific decomposed expression is as follows:
ESD＝(∑A C +∑A b +∑A O ) 2
in the formula, omega C ,Ω B ,Ω O Respectively representing the reflected path from the chest, from other parts of the user's body and from objects in the environment, omega C ∪Ω B ∪Ω O ＝Ω I ，A c Representing the reflected signal from the chest, A b Representing reflected signals from other parts of the user's body, A o Representing signals reflected by objects in the environment.
S2: extracting the mixed signal to obtain a corresponding voice signal and a corresponding chest cavity action signal;
s3: extracting the characteristics of the voice signal and the thoracic action signal obtained in the step S2;
s4: and performing verification by using the characteristics obtained in the step S3.
In a specific embodiment, the step S2 specifically includes: and processing the mixed signal by using a high-pass filter to obtain a corresponding voice signal, and processing the mixed signal by using a low-pass filter to obtain a corresponding thoracic action signal.
In a specific embodiment, the step S3 specifically includes:
s31: establishing an existing background model according to the surrounding environment to process the voice signal, and smoothing the voice signal processed by the background model by using a sliding window to obtain a final voice command and an interval of the voice command so as to eliminate interference;
s32: establishing an adaptive filter to separate and reduce noise of the chest motion signal to obtain a final chest motion signal; an adaptive filter bank is established to decompose the mixed signal into a plurality of signals, so that the thoracic cavity action signal with fine granularity can be extracted while non-static interference is well eliminated.
In a specific embodiment, the step S4 includes:
s41: identifying the thoracic cavity motion signal processed in the step S32, determining whether a user is present, and if so, proceeding to step S42;
specifically, the FFT of the chest motion signal is calculated, and the peak value is selected within the frequency range extracted by the low-pass filter, and the upper and lower limits of the peak value may be set to 0.16Hz and 0.6Hz, i.e., the frequency range is 0.16-0.6Hz, which is consistent with the human respiratory frequency, and which may effectively shield the replay attack performed by the electronic speaker.
S42: combining a sound signal of a user with a chest cavity action signal to obtain more reliable authentication, firstly, synchronizing the sound signal of the user with the chest cavity action signal, segmenting the chest cavity action signal according to the interval of the voice command, and obtaining a correlation coefficient of the chest cavity action signal and the voice command;
wherein, the specific expression of the correlation coefficient is as follows:
Figure BDA0003332407650000051
wherein N is the number of samples.
S43: and extracting features from the correlation coefficient, and training a deep neural network by using the features for the identification of the user.
Specifically, the Mel frequency cepstrum coefficient containing 13-dimensional features and the 13-dimensional first-order difference and the 13-dimensional second-order difference are used as the features, so that a 39-dimensional feature vector is obtained and used for constructing a user identification classifier, and the method is favorable for better distinguishing single users.
Specifically, given the extracted features (39 vimel frequency cepstrum coefficients of the speech signal and corresponding chest motion, i.e., MFCC fusion features), a deep neural network is trained for user authentication. However, since the number of samples for user registration is very rare, we propose to use meta-learning to complete the training of a neural network based on a priori knowledge. Specifically, the embodiment of the invention adopts a simple and effective meta-learning algorithm repeat to solve the problem of insufficient training samples. The principle of replay is to initialize a neural network which can adapt to a new task quickly, so that the initialized neural network has good generalization performance, then starting from the initialized neural network, fine tuning is carried out on network parameters by using some new samples, the problem that only a small number of training samples exist during the registration of a new user can be solved, the identification of the user is realized by using a deep neural network, a simple and effective meta-learning model is adopted to solve the problem that the training samples are insufficient,
the neural network consists of four convolutional layers, each convolutional layer is provided with 4 x 4 two-dimensional convolutional kernels, then a linear rectifying layer, a pooling layer and finally a full-connection layer, and a Sigmoid function is used for detecting whether a current input sample is from a legal user, if so, the login is successfully allowed, and otherwise, the login is refused.
Referring to fig. 2, an embodiment of the present invention further provides an intelligent voice authentication system, including:
the data acquisition module 1 is used for acquiring voice signals and thoracic cavity action signals of a user;
the data analysis module 2 is connected with the data acquisition module 1 and is used for analyzing and processing the voice signal and the thoracic cavity action signal;
the feature extraction module 3 is connected with the data analysis module 2, and is used for extracting features of the voice signal and the chest cavity action signal which are analyzed and processed;
and the intelligent terminal 4 is connected with the feature extraction module 3 and used for verifying according to the extracted features.
In a specific embodiment, the data analysis module 2 includes:
a first data processing unit 21, the first data processing unit 21 being configured to process the voice signal;
a second data processing unit 22, said second data processing unit 22 being adapted to process said chest movement signal.
In a specific embodiment, the feature extraction module 3 includes:
a voice signal extracting unit 31, configured to extract the processed voice signal to obtain a voice instruction;
and a thoracic action signal extraction unit 32, configured to extract the processed voice signal to obtain a thoracic action signal.
Further, an embodiment of the present invention further provides a computer-readable storage medium, on which computer instructions are stored, and when the computer instructions are executed, the method according to any one of the above embodiments is implemented.
To verify the effect of the above method, in a simple replay attack, the attacker only records the voice command signal of the attacked user, i.e. the distance between the attacker device and the victim device exceeds 5m, to ensure that the attacker can only record the voice command signal.
Three users are randomly selected as the attacked, five volunteers serve as imitators, and the record is played to the smartphone of the attacked for 20 times at a distance of about 15cm by using the smartphone. In an advanced replay attack, the attacker records both the voice command signal and the corresponding chest movement signal of the attacked user. The distance between the attacker's device and the victim device is about 15cm to ensure that the attacker can record both the voice command signal and the chest movement signal of the attacked user, other experimental settings being the same as for a simple replay attack. The experimental procedure allowed modelers to practice before the attack, each modeler would see the chest motion signals of the attacked user and their own chest motion signals on the smartphone screen, so that the modelers could improve their modelling, each volunteer modelled each attacked 20 times, the results showed that the modelling attack was slightly more effective than the gravity-cast attack, but the probability of successfully identifying the modelling attacker was still as high as 98.77%
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (7)

1. An intelligent voice authentication method, comprising:
s1: collecting a mixed signal sent by a user;
the mixed signal comprises signals reflected by the chest, signals reflected by other parts of the body and signals reflected by objects in the environment, and the specific expression is as follows:
ESD＝(∑A C +∑A b +∑A O ) 2
in the formula, A c Representing the reflected signal from the chest, A b Representing reflected signals from other parts of the user's body, A o A signal representing reflections from objects in the environment;
s2: extracting the mixed signal to obtain a corresponding voice signal and a corresponding chest cavity action signal;
s3: extracting the characteristics of the voice signal and the thoracic cavity action signal obtained in the step S2;
s4: performing verification by using the characteristics obtained in the step S3;
the step S4 includes:
s41: identifying the chest cavity action signals subjected to feature extraction, judging whether a user exists, if so, entering step S42, and the specific processing procedure comprises the following steps:
firstly, calculating FFT of the thoracic cavity motion signal, and selecting a peak value within a frequency range extracted by a low-pass filter, wherein the peak value range is 0.16-0.6 Hz;
s42: according to the interval of the voice commands subjected to feature extraction, synchronizing a sound signal of a user with a thoracic cavity action signal, segmenting the thoracic cavity action signal, and acquiring a correlation coefficient of the thoracic cavity action signal and the voice commands;
s43: extracting features from the correlation coefficient, and realizing the identification of the user by using a feature training deep neural network, wherein the specific processing process comprises the following steps:
the features are a Mel frequency cepstrum coefficient containing 13-dimensional features, a 13-dimensional first-order difference and a 13-dimensional second-order difference of the Mel frequency cepstrum coefficient, corresponding 39-dimensional feature vectors are obtained, and the 39-dimensional feature vectors are used for training a deep neural network to achieve user authentication.
2. The intelligent voice authentication method according to claim 1, wherein the step S2 specifically includes: and processing the mixed signal by using a high-pass filter to obtain a corresponding voice signal, and processing the mixed signal by using a low-pass filter to obtain a corresponding thoracic action signal.
3. The intelligent voice authentication method according to claim 2, wherein the step S3 specifically includes:
s31: establishing a background model to process the voice signal, and smoothing the voice signal processed by the background model by using a sliding window to obtain a final voice command and a voice command interval;
s32: and establishing an adaptive filter to separate and reduce noise of the chest motion signal to obtain a final chest motion signal.
4. An intelligent voice authentication system using the intelligent voice authentication method according to any one of claims 1 to 3, comprising:
the data acquisition module (1) is used for acquiring a voice signal and a chest cavity action signal of a user;
the data analysis module (2), the data analysis module (2) is connected with the data acquisition module (1) and is used for analyzing and processing the voice signal and the thoracic cavity action signal;
the characteristic extraction module (3) is connected with the data analysis module (2) and is used for extracting the characteristics of the voice signal and the thoracic cavity action signal which are analyzed and processed;
and the intelligent terminal (4) is connected with the feature extraction module (3) and used for verifying according to the extracted features.
5. An intelligent voice authentication system according to claim 4, wherein the data analysis module (2) comprises:
a first data processing unit (21), the first data processing unit (21) being configured to process the speech signal;
a second data processing unit (22), the second data processing unit (22) being configured to process the chest action signal.
6. An intelligent voice authentication system according to claim 5, wherein the feature extraction module (3) comprises:
the voice signal extraction unit (31) is used for extracting the processed voice signal to obtain a voice instruction;
and the thoracic cavity action signal extraction unit (32) is used for extracting the processed voice signal to obtain a thoracic cavity action signal.
7. A computer-readable storage medium having computer instructions stored thereon which, when executed, implement the method of any one of claims 1-3.
CN202111284667.5A 2021-11-01 2021-11-01 Intelligent voice authentication method, system and storage medium Active CN114003885B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111284667.5A CN114003885B (en) 2021-11-01 2021-11-01 Intelligent voice authentication method, system and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111284667.5A CN114003885B (en) 2021-11-01 2021-11-01 Intelligent voice authentication method, system and storage medium

Publications (2)

Publication Number Publication Date
CN114003885A CN114003885A (en) 2022-02-01
CN114003885B true CN114003885B (en) 2022-08-26

Family

ID=79926354

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111284667.5A Active CN114003885B (en) 2021-11-01 2021-11-01 Intelligent voice authentication method, system and storage medium

Country Status (1)

Country Link
CN (1) CN114003885B (en)

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113069105A (en) * 2021-03-26 2021-07-06 北京理工大学 Method for detecting smoking behavior of driver by using loudspeaker and microphone of smart phone

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170258350A1 (en) * 2016-03-10 2017-09-14 Hong Wen FANG Heart Murmur Detection Device and Method Thereof
CN107391994A (en) * 2017-07-31 2017-11-24 东南大学 A kind of Windows login authentication system methods based on heart sound certification
CN107481736A (en) * 2017-08-14 2017-12-15 广东工业大学 A kind of vocal print identification authentication system and its certification and optimization method and system
CN110010133A (en) * 2019-03-06 2019-07-12 平安科技（深圳）有限公司 Vocal print detection method, device, equipment and storage medium based on short text
CN112017658A (en) * 2020-08-28 2020-12-01 北京计算机技术及应用研究所 Operation control system based on intelligent human-computer interaction
CN112927694B (en) * 2021-03-08 2022-09-13 中国地质大学（武汉） Voice instruction validity judging method based on fusion voiceprint features

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113069105A (en) * 2021-03-26 2021-07-06 北京理工大学 Method for detecting smoking behavior of driver by using loudspeaker and microphone of smart phone

Also Published As

Publication number Publication date
CN114003885A (en) 2022-02-01

Similar Documents

Publication Publication Date Title
CN108597505B (en) Voice recognition method and device and terminal equipment
CN105702263B (en) Speech playback detection method and device
CN108899044B (en) Voice signal processing method and device
CN110956957B (en) Training method and system of speech enhancement model
CN110503971A (en) Time-frequency mask neural network based estimation and Wave beam forming for speech processes
CN106599866A (en) Multidimensional user identity identification method
CN109711350B (en) Identity authentication method based on lip movement and voice fusion
CN112347450B (en) Identity verification method based on blink sound signal
CN111583936A (en) Intelligent voice elevator control method and device
Shang et al. Voice liveness detection for voice assistants using ear canal pressure
Alegre et al. Evasion and obfuscation in automatic speaker verification
CN114003885B (en) Intelligent voice authentication method, system and storage medium
CN116312559A (en) Training method of cross-channel voiceprint recognition model, voiceprint recognition method and device
Gofman et al. Hidden markov models for feature-level fusion of biometrics on mobile devices
CN114333844A (en) Voiceprint recognition method, voiceprint recognition device, voiceprint recognition medium and voiceprint recognition equipment
CN111563244A (en) Identity authentication method, identity authentication device, computer equipment and storage medium
CN110020520A (en) A kind of recognition of face assistant authentification method and system based on voice signal
CN113626785B (en) Fingerprint authentication security enhancement method and system based on user fingerprint pressing behavior
CN115348049A (en) User identity authentication method using earphone inward microphone
WO2022156562A1 (en) Object recognition method and apparatus based on ultrasonic echo, and storage medium
CN110971755B (en) Double-factor identity authentication method based on PIN code and pressure code
CN116910732A (en) User non-sense trusted identity authentication method, system and terminal integrating keystroke and sound signals
CN110992980B (en) Hidden latent channel identification method based on edge calculation
Guo et al. PalmEcho: Multimodal Authentication for Smartwatch via Beating Gestures
US20210319803A1 (en) Methods and techniques to identify suspicious activity based on ultrasonic signatures

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant