A Method for Improved Dual-Channel Blind Signal Separation via Multi-Source Activity Detection

Abstract

The invention discloses a method for improving dual-channel blind signal separation through multi-sound source activity detection. This method is based on the dual-channel TRINICON algorithm for blind signal separation, and compares the power before and after the preliminary processing. If the signal power of one output channel is significantly lower than that of the other output channel, the target sound source to be suppressed in this signal can be judged. In the active state, according to this, it can be judged whether each target sound source in each segment of data is in the active state. The TRINICON algorithm is corrected by using the results of multi-sound source activity judgment, and the data of the target sound source activity is used to update the filter coefficients, so as to achieve the purpose of removing interference and improving the performance of speech separation. The method of the invention can effectively improve the separation performance of the TRINICON method in the scene of discontinuous interleaved mixing and sparse mixing.

Description

technical field [0001] The present invention relates to the technical field of blind signal separation, in particular to a method for improving the performance of blind signal separation through multi-sound source activity detection based on a dual-channel system. An algorithm for multi-sound source activity detection is added to the signal separation process. Background technique [0002] The problem of separating convolutional mixtures of unknown time series has important applications in several domains. An important example is the so-called cocktail party problem, where a single speech signal is extracted from the mixture of multiple speakers in a reverberant acoustic environment. Due to reverberation, the source signal for this separation problem is filtered through a linear multiple-input multiple-output (MIMO) system before being picked up by the microphone array. Blind Signal Separation (BSS) is an algorithm that uses a microphone array, does not require prior infor...

AI Technical Summary

Problems solved by technology

In practical applications, speech signals are sometimes intermittently detected or mixed sparsely, and the offline algorithm does not consider the active state of the speech, and counts each segment of speech equally in the calculation, which will lead to performance degradation, especially when Non-causal filters need to be generated when the individual sound sources are on one side of the midline of the microphone array

Images