Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Voiceprint identification method

A technology of voiceprint recognition and variance, which is applied in speech analysis, instruments, etc., can solve the problem that the Gaussian mixture model cannot satisfy incremental learning, and achieve the effect of increasing the computational burden, reducing training time, and reducing the feature dimension

Active Publication Date: 2012-12-12
NANJING UNIV
View PDF3 Cites 90 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Ordinary Gaussian mixture models cannot meet the requirements of incremental learning, and have to relearn all the data each time under the requirements of intermittent learning

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 identification method
  • Voiceprint identification method
  • Voiceprint identification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0034] The invention discloses a voiceprint recognition method, comprising the following steps:

[0035] Step 1, preprocessing the segmented speech data of each speaker in the training speech set to form a set of sample sets corresponding to each speaker after preprocessing, and the speech data of a speaker uniquely corresponds to a sample set;

[0036] Step 2, extracting Mel cepstral coefficients for each sample in all sample sets;

[0037] Step 3: Select a sample set one by one and randomly select the Mel cepstrum coefficients of some of the samples, train the Gaussian mixture model on the sample set until all the sample sets are trained to obtain the Gaussian mixture model, and combine all the Gaussian mixture models into one model library;

[0038] Step 4, performing incremental learning on the samples that have not been selected for training in step 3 and the Gaussian mixture models of their corresponding sample sets one by one to obtain all optimized Gaussian mixture mo...

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 voiceprint identification method. The voiceprint identification method comprises the following steps of: 1, preprocessing segmented speech data of each speaker in a training speech set to form a group of sample sets corresponding to each speaker; 2, extracting Mel-frequency cepstrum coefficients from each sample in all sample sets; 3, selecting a sample set one by one and randomly selecting the Mel-frequency cepstrum coefficients of part samples of the sample set, and training a Gaussian mixture model for the sample set; 4, performing incremental learning on the samples which are not selected and trained in the step 3 and the Gaussian mixture model of the sample set corresponding to the sample set one by one to obtain all optimized Gaussian mixture models, and optimizing a model library by utilizing all optimized Gaussian mixture models; and 5, inputting and identifying test voice data, identifying the Gaussian mixture model of the sample set corresponding to the test voice data by utilizing the optimized model library in the step 4, and adding the test voice data to the sample set corresponding to the speaker.

Description

technical field [0001] The invention relates to the technical field of speech feature extraction and recognition in the field of biological feature recognition, in particular to a voiceprint recognition method. Background technique [0002] Biometric identification technology refers to the technology that uses the physiological or behavioral characteristics that human beings possess to identify their identity for identity verification. Compared with traditional identity verification technology, biometric technology can provide more convenient user services, provide higher security level, reliability, and is increasingly used in identity authentication of modern security systems. [0003] The so-called voiceprint (Voiceprint) is the sound wave spectrum that carries speech information displayed by electroacoustic instruments. Modern scientific research shows that voiceprint is not only specific, but also relatively stable. After adulthood, the human voice can remain relative...

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): G10L17/00
Inventor 申富饶唐泽林赵金熙程佳
Owner NANJING 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
Eureka Blog
Learn More
PatSnap group products