Unlock instant, AI-driven research and patent intelligence for your innovation.

Voice enhancement method and device

A voice enhancement and voice technology, applied in voice analysis, instruments, etc., to achieve effective noise masking and high voice quality

Active Publication Date: 2015-04-29
陈卓 +1
View PDF6 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At the same time, it is inevitable to be disturbed by noise

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
  • Voice enhancement method and device
  • Voice enhancement method and device
  • Voice enhancement method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0030] Such as figure 1 As shown, the speech enhancement method, the method includes:

[0031] S100. Collect two channels of noisy speech signals from different spatial positions; of course, there may be more than two channels;

[0032] S200, the S200 step is used to process the noisy speech signal obtained in step S100, the specific processing steps include parallel S210 and S220, which are all based on the noisy speech signal in S100, wherein S210 is Separation from the noisy speech signal to obtain a preliminary enhanced speech signal, and further processing, S220 is to separate the noise from the noisy speech signal to obtain a noise reference signal; for the convenience of description, the steps of S210 are further refined into S211, S212 and S213, further subdivided into S221, S222, S223 and S224 in step S220, specifically as follows:

[0033] S210：

[0034] S211: Perform preliminary speech enhancement on the noisy speech signal obtained in step S100 to obtain a prel...

Embodiment 2

[0073] Speech enhancement device, described device comprises:

[0074] Speech collection module: used to collect multiple noisy speech signals from multiple different spatial positions, specifically a microphone can be used;

[0075] Speech enhancement module: used to obtain a preliminary enhanced speech signal, specifically a fixed beamformer can be used;

[0076] The first human ear filtering module: used to perform frequency division and windowing frame-by-frame processing on the preliminary enhanced voice signal output in the voice enhancement module to form N voice T-F units;

[0077] The first feature extraction module: for extracting the idiosyncratic value of each speech T-F unit;

[0078] Noise reference module: including a second human ear filter module, a second feature extraction module, a second feature value comparison module and a second masking module; specifically, frequency division and addition of noisy speech signals by the second human ear filter module ...

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 voice enhancement method and device. The method comprises the following steps: acquiring multi-channel noisy voice signals from a plurality of different space positions, obtaining preliminary-enhanced voice signals and noise reference signals according to the acquired noisy voice signals, performing the contrast of characteristic values on the preliminary enhanced voice signals and the noise reference signals, performing masking processing on the preliminary-enhanced voice signals based on a contrasted ratio to obtain final enhanced voice signals for output. According to the voice enhancement method and device, the 360-degree comprehensive coverage of the voice signals is realized, the signal-to-noise ratio is improved by 5-10 dB in comparison with that of a traditional noise suppression method (generalized side lobe canceling), and under the multi-noise environment, the noise can be more efficiently masked, and a voice signal of a target sound source can be enhanced, so that the higher voice quality is ensured.

Description

[technical field] [0001] The invention relates to the field of speech enhancement, in particular to a speech enhancement method and device. [Background technique] [0002] Users need to face a variety of complex environments when using hearing devices. The target voice is changing at any time, such as sometimes it is human voice, sometimes it is music, and the intensity and direction of the target voice are all changing at any time. At the same time, it is unavoidable to be disturbed by noise. [0003] The types of noise in different scenes vary widely. It may be steady state noise, it may be dynamic noise, it may be impact noise, it may be wind noise. The noise can be a single frequency tone, narrowband noise, or even wideband noise. [0004] Moreover, different users have different requirements for the target voice in different scenarios. For example, in a relatively quiet environment, the user needs clearer voice. In a relatively noisy environment, it is necessary to ...

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 Applications(China)
IPC IPC(8): G10L21/0208
Inventor 陈卓席金苗
Owner 陈卓