Method and device for suppressing noise of voice signals
A noise suppression and speech signal technology, applied in the field of speech signal noise suppression method and device, can solve the problems of reducing the system recognition rate, affecting the integrity of speech features, destroying the speech spectrum structure, etc.
Inactive Publication Date: 2013-09-11
BEIJING BAIDU NETCOM SCI & TECH CO LTD
View PDF7 Cites 9 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, the original purpose of existing speech enhancement algorithms is to enhance the subjective auditory experience of speech, but it destroys the spectral structure of speech and affects the integrity of speech features.
Therefore, directly applying the existing speech enhancement technology to the speech recognition system may not necessarily obtain positive benefits, and in many cases will reduce the recognition rate of the 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
[0049] figure 1Shown is a typical speech recognition system in the prior art, which is composed of noise suppression module, noise addition training module, recognition module and other parts. In the training phase, the noisy speech passes through the noise suppression device to obtain an enhanced speech signal, and the noise-adding training module uses the signal to perform noise-adding model training to obtain a noise-adding acoustic model. In the recognition stage, the noisy speech signal is also passed through the noise suppression module, and the enhanced speech obtained is sent to the recognition module, and the recognition process is performed according to the pre-trained noise addition model and language model, and the final speech recognition result is obtained.
[0050] It can be seen that the noise suppression module is an important part of the entire speech recognition system. The existing noise reduction processing idea is to use the existing speech enhancement a...
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 discloses a method and a device for suppressing noise of voice signals. The method for suppressing the noise of the voice signals includes performing Mel-domain decomposition on a target voice signal X(f) to obtain K Mel-domain sub-band signals X(f); determining local minimum energy values of various sub-bands in a time domain and signals corresponding to the minimum energy values by the aid of preset time windows, estimating corresponding sub-band noise signals of the various sub-bands and acquiring an estimated signal N(f) of the noise in a full frequency band by means of splicing estimation results; using the N(f) as a removal target to generate a filter, and suppressing the noise of the target voice signal X(f). The i can be 1, or 2, ..., or K. According to the scheme, the method and the device have the advantages that the integrity of voice characteristics can be kept to the greatest extent in a noise reduction procedure, and accordingly the voice recognition accuracy is improved.
Description
technical field [0001] The invention relates to the technical field of audio processing, in particular to a method and device for suppressing noise of speech signals. Background technique [0002] With the development of the Internet, the objects that users search on the Internet are not limited to text content, and pictures, audio, video, etc. have all become objects supported by search engines. Among them, voice-based search has become a popular form of application. Different from traditional text search, noise is unavoidable in the input process of voice signal, so how to effectively reduce noise becomes the key to improve the performance of voice search. [0003] At present, the noise reduction processing ideas for voice search are mainly divided into two categories: the first category is to add noise to the pure voice feature signal in the feature library, and train the noisy voice feature data, thereby increasing the feature library data and user The matching degree ...
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 Applications(China)
IPC IPC(8): G10L21/0208G10L21/0232
Inventor 宋辉
Owner BEIJING BAIDU NETCOM SCI & TECH CO LTD