Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Voice spoofing attack detection method based on sound field and field pattern

A technology of deception attack and detection method, applied in speech analysis, instruments, etc., can solve the problems of low detection accuracy, difficult application and high cost

Inactive Publication Date: 2020-06-05
ZHEJIANG UNIV
View PDF1 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this type of detection method usually has low detection accuracy and is difficult to apply after the attack method and equipment are upgraded.
In addition, there is a defense method for liveness detection by wearing additional equipment on the person, which requires high cost and poor personal experience due to the need for additional equipment
At present, there is no voice spoofing attack detection method based on sound field and field pattern

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Voice spoofing attack detection method based on sound field and field pattern
  • Voice spoofing attack detection method based on sound field and field pattern
  • Voice spoofing attack detection method based on sound field and field pattern

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0095] Such as figure 2 As shown, in the embodiment, the sound emitted by the speaker is used as the counterfeit object for detection and authentication, and the voice signal is collected by using the actual vocal cords of the individual's human vocalizing body to emit sound, and the sound emitted by the actual vocal cords of the individual's human vocalizing body recorded by the speaker is played as the sounding body collection The voice signal is then detected and judged and compared according to the method of the present invention.

[0096] Figure 4 Shows the field pattern comparison between the real person and the speaker, and the original field pattern before dimensionality reduction is shown in the figure. The sound played by the speakers is the replay of the voice of individual 1, and the voice content of the five individuals is also the same.

[0097] Depend on Figure 4 It can be seen from the figure above that there is a significant difference between the field ...

Embodiment 2

[0099] In this embodiment, the living body authentication voices of a total of 20 people (14 men and 6 women) were collected, and voice spoofing attacks were carried out with 8 different types and grades of loudspeakers. Setting the field pattern feature dimension to 9 and selecting the frequency range of 0-4kHz, the detection accuracy rate for real individuals is 99.03%, and the detection accuracy rate for spoofing attacks is 99.18%.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a voice spoofing attack detection method based on a sound field and a field pattern. The method comprises the following steps: collecting voice through double microphones, extracting to obtain double-track voice signals, respectively processing the double-track voice signals, and extracting field lines; collecting and processing to obtain field patterns of a plurality of voices of the same person, obtaining field pattern distribution of the person, and training by using a Gaussian mixture model; repeating the above steps on the to-be-detected voice of the unknown sounder to calculate the field pattern of the to-be-detected voice, processing the to-be-detected voice by using the trained Gaussian mixture model to obtain a similarity probability, and judging and detecting whether the to-be-detected voice is a voice spoofing attack or not according to the similarity probability. According to the invention, the spoofing attack for the voice authentication system canbe accurately and effectively detected.

Description

technical field [0001] The invention belongs to a sounding body authentication detection method in the field of voice authentication technology and security technology, and in particular relates to a processing method capable of detecting deception attacks aimed at a voice authentication system. Background technique [0002] The voice authentication system is a security authentication system that identifies the speaker's identity by extracting the speaker's voice features and performing feature pattern learning and matching. Due to its low hardware requirements (only microphones are required), low cost, simple and convenient personal operation, and long-distance non-contact authentication, it has gradually become a mainstream personal authentication and access control method. Speakers, smart home and other devices have been widely used. [0003] However, existing voice authentication systems are generally vulnerable to voice spoofing attacks. Voice spoofing attack refers t...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G10L17/00G10L17/02G10L17/04G10L25/27G10L25/78
CPCG10L17/04G10L17/02G10L25/27G10L25/78
Inventor 徐文渊闫琛冀晓宇龙颜
Owner ZHEJIANG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com