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

Voice recognition method and system based on Fisher hybrid features and neural network

A hybrid feature, voice identification technology, applied in speech recognition, voice analysis, instruments, etc., can solve the problem of low accuracy of the voice identification system, and achieve the effect of improving the accuracy

Inactive Publication Date: 2020-02-11
HEFEI UNIV OF TECH
View PDF4 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] Aiming at the deficiencies in the prior art, the present invention provides a voice identification method and system based on Fisher mixed features and neural network, which solves the technical problem of low accuracy of the existing voice identification system

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 recognition method and system based on Fisher hybrid features and neural network
  • Voice recognition method and system based on Fisher hybrid features and neural network
  • Voice recognition method and system based on Fisher hybrid features and neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0094] Example 1: Using MFCC and CQCC alone to train the LSTM-GRU neural network, the accuracy rates obtained are 97.64% and 97.48% respectively;

Embodiment 2

[0095] Embodiment 2: Using MFCC-CQCC mixed feature training to feed into the same neural network as Embodiment 1, the accuracy rate obtained is 98.27%.

Embodiment 3

[0096] Embodiment 3: Test of resistance to mp3 compression: The samples used in Embodiment 1 and Embodiment 2 are all WAV format files, and 1000 samples are selected and compressed into mp3 format. Extract MFCC features, CQCC features and MFCC-CQCC mixed features respectively, and send them to the same neural network as in Embodiment 1 and Embodiment 2 respectively. The final accuracies are MFCC: 90.14%, CQCC: 60.64%, and MFCC-CQCC mixed features: 92.52%.

[0097] The embodiment of the present invention also provides a kind of speech identification system based on Fisher mixed feature and neural network, and above-mentioned system comprises computer, and above-mentioned computer comprises:

[0098] at least one storage unit;

[0099] at least one processing unit;

[0100] Wherein, at least one instruction is stored in the at least one storage unit, and the at least one instruction is loaded and executed by the at least one processing unit to implement the following steps:

...

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 provides a voice recognition method and system based on Fisher hybrid features and a neural network, and relates to the technical field of voice recognition. The method comprises the following steps of firstly, obtaining a to-be-tested voice and a voice sample set including intelligent synthetic voice data and natural human voice database data, and then obtaining MFCC features and CQCC features of voice samples in the voice sample set; secondly, obtaining MFCC-CQCC hybrid features of the voice samples based on Fisher criteria, the MFCC features and the CQCC features; thirdly, obtaining a voice recognition model based on the hybrid features and the preset neural network; and finally, judging whether the to-be-tested voice is an intelligent synthetic voice or a natural human voice based on the voice recognition model. According to the voice recognition method and system, in voice feature selection, single features are not selected and the MFCC-CQCC hybrid features based onthe Fisher criteria are selected; the hybrid features organically combine the MFCC and CQCC features, so that the voice synthesized by multiple algorithms can be effectively recognized; and the neuralnetwork is trained by using the hybrid features to obtain the voice recognition model, so that the accuracy of the voice recognition model can be effectively improved.

Description

technical field [0001] The invention relates to the technical field of voice recognition, in particular to a voice identification method and system based on Fisher mixed features and a neural network. Background technique [0002] With the continuous development of speech signal processing technology, the identity authentication system using the speaker's speech signal has been widely used in many industries. There are relatively large potential safety hazards in identity authentication using the speaker's voice signal, and the potential security risks include using synthetic voice to impersonate the speaker's voice. Therefore, how to distinguish between synthetic speech and natural human voice is the key to eliminating potential safety hazards. [0003] In the prior art, a common voice identification system utilizes voice features to identify whether the voice to be tested is a synthetic voice or a natural human voice. Speech features mainly include MFCC features and CQCC...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G10L15/02G10L15/16G10L25/03G10L25/24
CPCG10L15/02G10L15/16G10L25/03G10L25/24
Inventor 苏兆品季仁杰葛昭旭陈清郑宁军李顺宇张国富岳峰
Owner HEFEI UNIV OF TECH
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