Frequency-domain blind separation sequencing algorithm of convolutive speech signals

A speech signal and sorting algorithm technology, applied in speech analysis, speech recognition, complex mathematical operations, etc., can solve problems such as order uncertainty, and achieve the effect of good robustness and accuracy

Inactive Publication Date: 2011-11-02
SHANDONG UNIV
View PDF3 Cites 22 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The present invention aims at the deficiencies of existing methods for solving the problem of order uncertainty in convolution frequency domain

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
  • Frequency-domain blind separation sequencing algorithm of convolutive speech signals
  • Frequency-domain blind separation sequencing algorithm of convolutive speech signals
  • Frequency-domain blind separation sequencing algorithm of convolutive speech signals

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] What the present invention uses is the convolution aliasing model of K*K (K source signals, K observation signals): Where the observed signal vector x(n)=[x 1 (n),x 2 (n),...,x K (n)] T , source signal vector s(n)=[s 1 (n), s 2 (n), .., s K (n)] T (The superscript "T" means transpose), N is the length of the FIR filter, is the K×K hybrid filter matrix at delay l, where h ij is the impulse response from the jth source signal to the ith sensor. For convolutional blind separation, the goal is to find L K×K separation filter matrices W(l) to estimate the source signal After the formula is short-time Fourier transformed (STFT), the convolutional aliasing model is converted into instantaneous aliasing in each frequency band, that is, in the frequency band f k , there is Y(f k ,τ)=W(f k )X(f k , τ). Through the frequency domain ICA (Independent Component Analysis) algorithm, a K×K separation matrix W(f k ). W(f k ) Each row is an estimated vector of differe...

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 frequency-domain blind separation sequencing algorithm of convolutive speech signals, and is characterized by comprising the following steps: converting time domains of convolutive overlapping speech signals into frequency domains of the convolutive overlapping speech signals, and using a frequency-domain ICA (independent component analysis) algorithm to carry out blind separation on each frequency band; and carrying out sequencing by using the sequencing algorithm: (1) selecting standard frequency bands to carry out alignment; (2) sequencing remaining frequency bandsin accordance with the sequenced standard frequency bands; and (3) marking frequency bands which generate sequencing errors possibly, and using DOA (direction of arrival) evaluation based on separation matrixes to carry out complementary alignment. In the algorithm, the thoughts on standard frequency band alignment are combined with the advantages of a sequencing algorithm based on frequency bandcoherence and a sequencing algorithm based on the DOA, thus the algorithm provided by the invention has the advantage of good robustness and accuracy and is still applied in real environments.

Description

technical field [0001] The invention relates to a method for solving the order uncertainty problem existing in the frequency domain blind source separation process of a convolution mixed speech signal, belonging to the field of speech signal processing. Background technique [0002] The blind separation algorithm is a method to estimate the original signal from the observed mixed signal when the source signal and the mixing process are unknown. Blind separation algorithm is widely used in speech signal processing, wireless communication, medical signal processing and many other fields. For the convolution blind separation problem, there are currently two types of algorithms: the first type is the time domain separation algorithm, and the second type is the frequency domain separation algorithm. Compared with the blind source separation algorithm in the time domain, the blind separation algorithm of speech signal in frequency domain has attracted the attention of researchers...

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): G10L21/02G10L15/22G06F17/15G10L21/0232G10L21/0272G10L25/18
Inventor 刘琚王倩杜军刘朝晨吕宁
Owner SHANDONG 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