Voiceprint recognition method, electronic device and computer-readable storage medium

A voiceprint recognition and computer program technology, applied in the electronic field, can solve the problem of low accuracy of voiceprint recognition and achieve the effect of real-time evaluation of false positive rate

Active Publication Date: 2020-08-18
SHENZHEN UNIV
View PDF13 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Due to the complexity of the actual situation, different types of recognition subsystems in the prior art may not necessarily adapt to the initial weights. Therefore, the method of using fixed weights makes the accuracy of voiceprint recognition not high.

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
  • Voiceprint recognition method, electronic device and computer-readable storage medium
  • Voiceprint recognition method, electronic device and computer-readable storage medium
  • Voiceprint recognition method, electronic device and computer-readable storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0057] The embodiment of this application provides a voiceprint recognition method, please refer to Figure 1-a , the voiceprint recognition method mainly includes the following steps:

[0058] 101. Obtain voice data to be analyzed;

[0059] The embodiment of the present invention is applied to a voiceprint recognition system, and the voiceprint recognition system includes K subsystems, where K is an integer greater than zero. The system architecture of the voiceprint recognition system in the embodiment of the present invention can refer to Figure 1-b .

[0060] Wherein, each subsystem in the voiceprint recognition system can correspond to different types of voiceprint recognition, and the types of voiceprint recognition include: emotion recognition, age recognition, and language recognition. Furthermore, each subsystem can also correspond to each subcategory in a recognition scenario. For example, in speech recognition, a subsystem corresponds to a language (such as Chine...

Embodiment 2

[0097] In the embodiment of the present invention, error-prone classifiers need to be constructed, please refer to Figure 1-c Methods include:

[0098] 201. Establish a training database;

[0099] Take the short-term speech data set as the test data set of each subsystem, mark all misjudged speech segments in the test process as N different labels according to different subsystems, and use it as a training database, and the N is an integer greater than zero .

[0100] 202. Extracting MFCC Mel frequency cepstral coefficient features;

[0101] For each piece of short-term speech data in the training database, Mel Frequency Cepstrum Coefficient (MFCC) features are extracted.

[0102] 203. Train the overall change matrix;

[0103]According to the extracted MFCC features, the universal background model (Universal Background Model, UBM) is trained, and the overall change matrix T is trained.

[0104] 204. Obtain the change factor feature of the short-term voice data;

[0105]...

Embodiment 3

[0115] In the embodiment of the present invention, taking the hybrid system of language recognition as an example, the voiceprint recognition method in the embodiment of the present invention is described in detail, including:

[0116] 1. For the architecture of the hybrid system for language recognition in the embodiment of the present invention, please refer to Figure 1-b , each subsystem independently gives the probability values ​​of N different languages.

[0117] 2. Let x be an input voice, and the output of each subsystem is shown in the following table:

[0118] Language code Language 1 Language 2 … LanguageN Language Probability P(L 1 |x)

P(L 2 |x)

… P(L N |x)

[0119] P f (L j |x)(i=1,2,…,N) Each subsystem independently gives a certain input voice belonging to a certain language L j (j=1,2,...,N) probability, and the sum of all probabilities is also 1, namely:

[0120]

[0121] Arrange the probabilities of all lan...

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

A voiceprint recognition method, an electronic device, and a computer readable storage medium. The voiceprint recognition method comprises: acquiring voice data to be analyzed; extracting a change factor feature in the voice data; using an fallible point classifier to perform miscalculation classification on the voice data according to the change factor feature so as to obtain relative miscalculation probabilities of voice data miscalculation in K subsystems; determining the offset of the relative miscalculation probability corresponding to any subsystem with respect to the average relative miscalculation probability of the K subsystems, and calculating a final fusion weight of the corresponding subsystem according to the offset; and weighting the recognition result of each subsystem by means of the final fusion weight, and obtaining a comprehensive recognition result of the voice data according to the recognition result of each subsystem after weighting.

Description

technical field [0001] The present application relates to the field of electronic technology, and in particular to a voiceprint recognition method, an electronic device, and a computer-readable storage medium. Background technique [0002] With the popularity of smart devices and related hardware facilities, voice interaction has become an indispensable part of human-computer interaction. There are more and more application scenarios related to voiceprint in voice interaction, including but not limited to: voiceprint attendance check-in, software login, bank transfer and account opening verification, wake-up of virtual voice assistant, personalized interaction for different user groups Wait, these systems all use voiceprint without exception. The so-called voiceprint refers to the unique voice characteristics of each person. In real life, everyone's voice has its own characteristics when speaking. Generally speaking, voiceprint recognition is divided into the following ca...

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 Patents(China)
IPC IPC(8): G10L17/06G10L25/24G10L25/51
CPCG10L17/06G10L25/24G10L25/51
Inventor 郑能恒林吉
Owner SHENZHEN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap