Audio enhancement method and system

Abstract

The invention discloses an audio enhancement method. Spatial spectrum of the original multi-channel audio frequency is obtained through a direction of arrival (DOA) estimation algorithm. A plurality of peak values greater than a set threshold value are obtained from the spatial spectrum; and a plurality of estimation direction values of the plurality of peak values are obtained according to the DOA estimation method. A spatial covariance matrix of the plurality of estimation direction values is obtained according to the plurality of estimation direction values and a guiding vector of a microphone array. A CGMM (complex Gaussian mixture model) is initialized and established according to the spatial covariance matrix; and a parameter of the CGMM is updated iteratively by a clustering method.The original multi-channel audio frequency is enhanced by an MVDR (minimum variance distortionless response) beamforming algorithm to obtain an enhanced audio. The method reduces the number of timesthat an EM algorithm iteratively updates the parameter of the CGMM, thereby greatly reducing the amount of calculations. Meanwhile, a category of time-frequency point masking values obtained for eachfrequency band is determined, so that masking values of the same category for each frequency band can be combined together, to solve the problem of fuzzy sequencing.

Description

technical field [0001] The invention belongs to the technical field of speech recognition, in particular to an audio enhancement method and system. Background technique [0002] At present, the masking value of the time-frequency point is mostly obtained through CGMM (complex Gaussian mixture model, complex Gaussian mixture model), and then MVDR (minimum variance distortionless response, minimum variance distortionless response) is used for speech enhancement. [0003] However, the defects of the above method mainly lie in the following two aspects. The first aspect is that after the parameters of the CGMM model are randomly initialized, in order to make the CGMM model achieve better results, it is usually necessary to use the EM algorithm to iteratively update the parameters more than 20 times, so the calculation of the algorithm The complexity is very large. The second defect is that since the algorithm is performed in the frequency domain, the calculations between the fr...

AI Technical Summary

Problems solved by technology

[0003] However, the defects of the above method mainly lie in the following two aspects. The first aspect is that after the CGMM model parameters are randomly initialized, in order to make the CGMM model achieve better results, it is usually necessary to use the EM algorithm to iteratively update the parameters more than 20 times, so the calculation of the algorithm very complex

The second defect is that since the algorithm is performed in the frequency domain, the calculations between the frequency bands are independent of each other

[0006] In the implementation process of the method in the prior art, in order to ensure the later application of audio, such as recognition and other operations, it is necessary to iterate the original collected audio multiple times, so the computational complexity of the algorithm is very large

The category corresponding to each masking value is uncertain, resulting in ambiguous sorting problems

Images