Single-channel voice enhancement method and device, storage medium and terminal

A voice enhancement, single-channel technology, applied in voice analysis, instruments, etc., can solve problems such as poor noise reduction effect, affecting voice call quality, unsatisfactory, etc., to ensure voice integrity, real-time estimation, and significant enhancement Effect

Active Publication Date: 2022-07-15
SPREADTRUM COMM (TIANJIN) INC
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] However, the performance of existing speech enhancement technologies in non-stationary noise and hands-free call scenarios is not satisfactory, and the noise reduction effect is poor, which seriously affects the quality of speech calls

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 enhancement method and device, storage medium and terminal
  • Single-channel voice enhancement method and device, storage medium and terminal
  • Single-channel voice enhancement method and device, storage medium and terminal

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0038] As mentioned in the background art, when people use mobile devices (such as mobile phones, telephone watches) to make daily calls, they are often exposed to a noisy background environment, and most of these noises are non-stationary noises in a statistical sense.

[0039] The traditional voice enhancement technology usually uses the Voice Activity Detection (VAD) method to judge whether there is voice in each frame of signal in the time domain, that is, to identify the voice frame and the pure noise frame from a noisy voice signal. The algorithm only estimates and updates the noise in the pure noise frame determined by VAD, and performs noise reduction in the speech frame according to the estimated noise spectrum.

[0040] This speech enhancement method is effective for stationary noise with little variation. However, for non-stationary noise, since the noise may change greatly in the speech frame, the noise spectrum estimated in the pure noise frame cannot truly reflec...

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

A single-channel voice enhancement method and device, storage medium, and terminal, the method comprising: acquiring a frequency domain amplitude spectrum of a current frame signal based on a received input signal; VAD processing is performed on the entire band to obtain the initial full-band amplitude spectrum gain function of the current frame signal; the full-band is divided into multiple sub-bands, and the current frame signal is based on the frequency domain amplitude spectrum of the current frame signal and the initial full-band amplitude spectrum gain function. The multiple sub-bands of the sub-bands are respectively subjected to VAD processing, and the initial full-band amplitude spectrum gain function is updated according to the VAD processing results of each sub-band to obtain the updated full-band amplitude spectrum gain function of the current frame signal; According to the frequency domain amplitude spectrum of the current frame signal And update the full-band amplitude spectrum gain function to calculate the spectrum after speech enhancement. The solution of the invention can effectively suppress non-stationary noise and protect the voice quality from loss, which is beneficial to improve the voice call quality of mobile devices such as mobile phones.

Description

technical field [0001] The present invention relates to the technical field of speech processing, in particular to a single-channel speech enhancement method and device, a storage medium and a terminal. Background technique [0002] With the popularization of mobile devices such as mobile phones and the construction and development of mobile networks, users have higher and higher requirements for the quality of voice calls. [0003] When making a voice call, the near-end speaker is often placed in a noisy background environment, and the noise in the environment will contaminate useful voice information. If the upstream voice signal containing noise is not processed, it will cause great trouble to the far-end receiver, making it impossible to accurately grasp the meaning of the voice. [0004] In addition, there are also cases where the near-end talker is not only in a noisy environment, but also turns on the hands-free calling mode during the call. For example, the driver ...

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 Patents(China)
IPC IPC(8): G10L21/0232G10L25/18
CPCG10L21/0232G10L25/18
Inventor 纪伟于伟维潘思伟雍雅琴董斐林福辉
Owner SPREADTRUM COMM (TIANJIN) INC
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