Single-channel voice separation method and device and electronic equipment

A speech separation, single-channel technology, applied in speech analysis, instruments, etc., to improve intelligibility, good separation effect, and reduce distortion rate

Active Publication Date: 2020-07-17
JIANGSU UNIV
View PDF7 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, temporal separation systems usually need to model extremely long input sequences, and need to mine th

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
  • Single-channel voice separation method and device and electronic equipment
  • Single-channel voice separation method and device and electronic equipment
  • Single-channel voice separation method and device and electronic equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0032] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts are within the protection scope of the present invention.

[0033] see figure 1 , a single-channel speech separation method based on a self-attention mechanism and a two-way recurrent neural network, the method includes the following steps:

[0034] Step 1, the encoder receives the mixed speech signal of multiple people speaking, and extracts the features of the mixed speech signal:

[0035] Using a one-dimensional convolutional neural network as an encoder to extract features X∈R from a mixed speech signal of multiple people speaking N×L , the feature is a 2-D tensor, where R represents a set of real numbers, L is the number o...

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 single-channel voice separation method and device and electronic equipment, and the method comprises the steps: extracting mixed voice signal features through an encoder, segmenting the extracted voice signal features and splicing the voice signal features into a 3-D tensor; modeling the spliced 3-D tensor by using a double-path recurrent neural network fused with a self-attention mechanism, and learning a long-time dependency relationship between voice signals; overlapping and adding the modeled 3-D tensors, and restoring the 3-D tensors into sequence voice signal features; and reconstructing the sequence voice signal features into pure voice signals by using a decoder to obtain separated voice signals. According to the invention, the long-time voice signals are modeled with the purpose of improving the voice separation performance, the long-time dependence relation between the voice signals is fully mined, the separation effect is good, the voice distortion rate is effectively reduced, and meanwhile, the intelligibility of voice separation is improved.

Description

technical field [0001] The invention relates to the fields of speech signal processing, pattern recognition and the like, and in particular to a single-channel speech separation method, device and electronic equipment. Background technique [0002] Single-channel speech separation refers to separating the pure speech of each person from a mixed speech of multiple people, which is an important branch in the field of signal processing. It has many practical applications in the real world, such as: separating pure speech signals from mixed noise speech to improve the accuracy of speech recognition and speaker identification. In the fields of video conference transcription, hearing assistance, mobile communication, etc., single-channel speech separation has broad application prospects and practical significance. [0003] Traditional single-channel speech separation mainly uses non-negative matrix factorization and auditory scene analysis. The non-negative matrix factorization ...

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): G10L21/0272G10L25/30
CPCG10L21/0272G10L25/30
Inventor 毛启容陈静静钱双庆高利剑
Owner JIANGSU 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