Speech de-noising method and speech de-noising device

A voice denoising, non-speech technology, applied in voice analysis, instruments, etc., can solve problems such as poor results

Active Publication Date: 2017-03-08
苏州谦问万答吧教育科技有限公司
View PDF4 Cites 31 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the noise in the speech part may be different from the noise in the non-speech part, especially if it is affec

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
  • Speech de-noising method and speech de-noising device
  • Speech de-noising method and speech de-noising device
  • Speech de-noising method and speech de-noising device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0023] figure 1 The flow chart of the voice denoising method provided in Embodiment 1 of the present invention, this embodiment can be used for voice denoising, the method can be executed by a voice denoising device, and the device can be implemented by software and / or hardware, the The device can be integrated into any intelligent terminal that provides voice denoising function. In specific implementation, the intelligent terminal can include: mobile terminals such as tablets, mobile phones, and e-readers. The above terminals are only examples, not exhaustive, including but not exhaustive. Limited to the smart terminals mentioned above.

[0024] see figure 1 , the method for speech denoising, comprising:

[0025] S110. Perform speech detection on the noisy speech signal to distinguish speech frames from non-speech frames.

[0026] The speech signal received by the smart terminal is a non-stationary time-varying noisy speech signal formed after being disturbed by the enviro...

Embodiment 2

[0044] figure 2 The flow chart of the speech denoising method provided by Embodiment 2 of the present invention. On the basis of the above embodiments, this embodiment proposes an effective speech feature combination when performing speech detection on noisy speech signals, which can more accurately distinguish Separate noisy speech frames and non-speech frames.

[0045] see figure 2 , the method for speech denoising, comprising:

[0046] S210. Extract speech features of the noisy speech signal.

[0047] The extracted noisy speech features include Mel cepstral coefficient MFCC, linear predictive coding residual and spectral centroid Centroid. Human beings have different perceptual abilities to different frequencies of speech: below 1kHz, it has a linear relationship with the frequency, and above 1kHz, it has a logarithmic relationship with the frequency. The higher the frequency, the worse the perception. In applications, only low-frequency MFCCs are often used, while m...

Embodiment 3

[0077] image 3 It is a flow chart of the speech denoising method provided by Embodiment 3 of the present invention. On the basis of the foregoing embodiments, this embodiment performs stationary noise suppression, non-speech noise suppression, and non-stationary noise suppression on noisy speech signals.

[0078] see image 3 , the method for speech denoising, comprising:

[0079] S310. Perform speech detection on the noisy speech signal to distinguish speech frames from non-speech frames.

[0080] S320. Perform noise estimation on the speech frame and the non-speech frame respectively, to obtain a fusion estimation value of the noise power spectrum.

[0081] S330. Perform stationary noise suppression, non-speech noise suppression, and non-stationary noise suppression on the noisy speech signal according to the fused estimated value of the noise power spectrum.

[0082] Exemplarily, after the fused estimated value of the noise power spectrum is obtained, denoising processi...

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

Embodiments of the invention disclose a speech de-noising method and a speech de-noising device. The method comprises the following steps: detecting the speech of a speech signal with noise to distinguish between speech frames and non-speech frames; estimating the noise of the speech frames and the noise of the non-speech frames to get a noise power spectrum fused estimated value, wherein the noise power spectrum fusion estimated value is the fused value of the noise power spectrum estimated value of the speech frames and the noise power spectrum estimated value of the non-speech frames; and de-noising the speech signal with noise according to the noise power spectrum fusion estimated value. According to the technical scheme provided by the embodiments of the invention, the noise of the speech frames and the noise of the non-speech frames are estimated, and the speech signal with noise is de-noised based on the noise estimation results of the speech frames and the non-speech frames. Thus, the de-noising effect of the existing speech de-noising scheme is improved effectively, and the quality of speech is improved.

Description

technical field [0001] Embodiments of the present invention relate to speech signal processing technologies, and in particular, to a speech denoising method and device. Background technique [0002] In the process of real-time voice communication, various noise interference problems will be encountered, especially for mobile devices such as mobile phones, the problem of voice noise is particularly prominent. In addition, in the case of playing sound through a speaker, due to the echo problem, the sound quality of speech in this case is easily affected by external environmental noise and nonlinear residual echo compared with remote recording. [0003] In order to improve the quality of voice communication, it is necessary to denoise the voice to improve the clarity of the voice. Traditional speech denoising algorithms usually assume that the noise is additive and stable, and use Voice Activity Detection (Voice Activity Detection, VAD) technology to distinguish noisy speech i...

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): G10L21/0216G10L25/78
Inventor 吴威麒张凯磊
Owner 苏州谦问万答吧教育科技有限公司
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