Single channel speech enhancement method based on attention-gated recurrent neural network

A recurrent neural network and speech enhancement technology, applied in speech analysis, instruments, etc., can solve problems such as poor non-stationary noise suppression effect, high computational complexity of speech enhancement methods, and inability to meet real-time requirements, achieving good application prospects, The effect of improving generalization ability and improving learning ability

Active Publication Date: 2019-08-02
NANJING INST OF TECH
View PDF15 Cites 25 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the problem that the existing speech enhancement method has a poor suppression effect on non-

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 speech enhancement method based on attention-gated recurrent neural network
  • Single channel speech enhancement method based on attention-gated recurrent neural network
  • Single channel speech enhancement method based on attention-gated recurrent neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0049] The present invention will be further described below in conjunction with the accompanying drawings.

[0050] Such as figure 1 As shown, the single-channel speech enhancement method based on the attention-gated recurrent neural network of the present invention includes the following steps.

[0051] Step (A), frame and window the noisy single-channel speech, extract 38-dimensional signal features, including Bark frequency cepstral coefficient and its derivative parameters, discrete cosine transform of pitch correlation coefficient, pitch period and spectral non-stationarity Measurement parameters, specifically including 18 Bark frequency cepstral coefficients, the first and second order time derivatives of the first 6 Bark frequency cepstral coefficients, the discrete cosine transform of the pitch correlation coefficients between the first 6 frequency bands, and 1 pitch period coefficient And a spectral non-stationarity measurement parameter;

[0052] Step (B), constru...

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 single-channel speech enhancement method based on an attention-gated recurrent neural network. The method comprises the steps of framing and winding single-channel speech with noise, and extracting 38-dimensional signal features; constructing a deep recurrent neural network for single-channel speech enhancement; constructing a training data set by using a pure speech library and a noise library; training the constructed deep recurrent neural network; inputting the extracted speech features with noise into the trained deep recurrent neural network, outputting a frequency band gain estimation value of the speech with noise, smoothing and interpolating to obtain interpolation gain; enabling the interpolation gain to act on the single-channel speech with noise so as to obtain an enhanced speech spectrum. The single-channel speech enhancement method can effectively inhibit the noise including nonstationary noise, and maintains low enough computation complexity, thus being used for real-time single-channel speech enhancement; the method is ingenious and novel in conception, thus having a good application prospect.

Description

technical field [0001] The invention relates to the technical field of speech enhancement, in particular to a single-channel speech enhancement method based on an attention-gated recurrent neural network. Background technique [0002] Speech enhancement, as a branch of speech signal processing, has important applications in speech communication, hearing aids, automatic speech recognition (ASR) system front-end and other fields. Speech enhancement is generally divided into single-channel speech enhancement and multi-channel speech enhancement. Single-channel speech enhancement is relatively more difficult to implement due to the lack of spatial information of the microphone array. [0003] Some unsupervised single-channel speech enhancement algorithms proposed earlier, such as spectral subtraction, Wiener filtering method, magnitude spectrum estimation based on minimum mean square error (MMSE) or spectral estimation method in logarithmic domain, due to the assumption of nois...

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): G10L21/0232G10L21/0264G10L25/18G10L25/24G10L25/30
CPCG10L21/0232G10L21/0264G10L25/18G10L25/24G10L25/30
Inventor 梁瑞宇孔凡留谢跃王青云程佳鸣孙世若赵力
Owner NANJING INST OF TECH
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