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

Identification method of speaker

A speaker recognition and speaker technology, applied in the field of speaker recognition, can solve problems such as channels and language outliers

Active Publication Date: 2016-01-20
INST OF ACOUSTICS CHINESE ACAD OF SCI +1
View PDF5 Cites 19 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to overcome the defect that the total variation factor in the existing speaker recognition method is multimodal in the overall distribution; and abnormal values ​​may appear due to the influence of channels, languages, etc., thereby providing a method that can effectively improve speaker recognition. System identification performance and speed methods

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
  • Identification method of speaker
  • Identification method of speaker
  • Identification method of speaker

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0066] Now in conjunction with accompanying drawing, the present invention is described in further detail:

[0067] refer to figure 1 , a process for generating a speaker recognition model includes:

[0068] Step 1-1), gather a certain amount of background speech data and target speaker speech data as training speech data, extract acoustic spectrum feature from described training speech data; This step comprises:

[0069] The training speech data of described step 1-1) is done front-end processing, and described training speech data front-end processing comprises cutting invalid speeches such as silence, music to training data, retains effective speech; Then extract from the training speech data through front-end processing The general-purpose Meier cepstrum feature (MFCC), and dynamically expand the feature to obtain the second-order difference cepstrum feature, so that each frame of the training speech data includes 60-dimensional feature vectors, and these feature vectors ...

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 relates to an identification method of a speaker. The method comprises: a speaker identification model is generated, background voices and target speaker voices are used as training data to obtain a first Gauss mixing-universal background model, a total changing space, a second Gauss mixing-universal background model, and a local linear discrimination analysis model; and a total changing factor of a to-be-identified voice and a posterior probability of the total changing factor are calculated based on the first Gauss mixing-universal background model, the total changing space, the second Gauss mixing-universal background model, and the local linear discrimination analysis model, the local linear discrimination analysis model is inputted to carry out conversion to obtain a low-dimension vector, and the vector is inputted into a rear-end identifier and an identification result is outputted. According to the invention, the discriminating property of speakers is enhanced; and the identification performance of the speaker is improved. Meanwhile, dimensionality reduction of the total changing factor is realized; the identification speed is enhanced; and the practicability is high.

Description

technical field [0001] The present invention relates to a method for identifying speaker information in voice data, more specifically, the present invention relates to a method for speaker identification based on local linear discriminant analysis. Background technique [0002] With the globalization of information in modern society, speaker recognition has become one of the research hotspots of speech recognition technology. Speaker recognition technology is a kind of identity verification technology --- biometric recognition technology. Compared with other identity verification technologies, speaker recognition is more convenient and natural, and has relatively low user intrusion. Speaker recognition tasks can be divided into speaker identification and speaker verification according to different types of practical applications. Among them, speaker identification is carried out within the range of all target speakers, and its performance is related to the number of target ...

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): G10L17/02
Inventor 周若华许云飞颜永红杨琳
Owner INST OF ACOUSTICS CHINESE ACAD OF SCI
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