A Noise Classification and Removal Method Based on Conference Scene
A noise and scene technology, applied in the field of noise classification and removal based on conference scenes, can solve problems affecting accuracy, reduce the robustness and anti-interference ability of the voiceprint recognition system, achieve noise removal, improve robustness and accuracy rate effect
Active Publication Date: 2020-12-15
NANJING UNIV OF POSTS & TELECOMM
View PDF7 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0002] Noise is everywhere. In the speaker recognition system, except for the voice of the target speaker, all external sounds are noise. If these noises are not removed, it will affect the accuracy of the voiceprint recognition system in judging the target speaker. , reducing the robustness and anti-interference ability of the voiceprint recognition system
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0025] The present invention will be described in further detail below in conjunction with the accompanying drawings and specific embodiments.
[0026] As shown in Figure 1, the described noise classification and removal method based on conference scene includes the following steps:
[0027] Step (1): Collect the noisy voice signal in the conference scene, confirm the noise source in the voice signal through manual identification, and then analyze the noise source introduced in each stage of the voiceprint recognition system. According to the noise source in the conference scene, The non-background noise in the conference scene is divided into non-target speaker noise, conference equipment noise, user behavior noise, and acquisition equipment noise;
[0028] Step (2): Use the control variable method to artificially create a single type of noise, collect the noisy speech signal under the condition of the noise, and obtain the time domain waveform and spectrogram of the speech s...
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
Login to View More Abstract
The invention provides a noise classified elimination method based on a conference scene. The noise classified elimination method comprises the following steps: step (1): firstly, dividing non-background noises in the conference scene into non-target speaker noises, conference equipment noises, user behavior noises and acquisition equipment noises according to characteristics of the noises under the conference scene; step (2): eliminating the acquisition equipment noises by utilizing a double-MIC noise reduction method according to the difference between different types of noises and target speaker noises in the aspect of a time-domain waveform graph and a speech spectrum; the double-MIC noise reduction method is used for removing the acquisition equipment noises and the user behavior noises are removed by utilizing a wavelet noise detection method; the conference equipment noises are removed by utilizing an SOX noise detection method and the non-target speaker noises are removed by utilizing a PCA (Principal Component Analysis) noise detection method, so that the noises under the conference scene are eliminated. The noise classified elimination method based on the conference scene, provided by the invention, has the advantages that the noises under the conference scene can be effectively removed and the robustness and accuracy of a voiceprint recognition system are greatly improved.
Description
technical field [0001] The invention relates to the technical field of intelligent conferences, in particular to a noise classification and removal method based on conference scenes. Background technique [0002] Noise is everywhere. In the speaker recognition system, except for the voice of the target speaker, all external sounds are noise. If these noises are not removed, it will affect the accuracy of the voiceprint recognition system in judging the target speaker. , reducing the robustness and anti-interference ability of the voiceprint recognition system. Contents of the invention [0003] The purpose of the present invention is to provide a noise classification and removal method based on a meeting scene that can effectively remove noise in a meeting scene and improve the robustness and accuracy of a voiceprint recognition system. [0004] In order to achieve the above object, the present invention adopts the following technical solution: a method for classifying an...
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
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G10L17/02G10L17/14G10L17/22G10L21/0216G10L21/0224
Inventor 张晖高财政赵海涛孙雁飞朱洪波
Owner NANJING UNIV OF POSTS & TELECOMM