Adaptive blind audio source extraction supervised by dominant speaker identification using x-vectors

TL;DR

This paper introduces an adaptive blind audio source extraction method that uses x-vectors for identifying the dominant speaker, improving extraction accuracy in dynamic and noisy environments.

Contribution

It presents a partially supervised algorithm combining independent vector analysis with x-vector based speaker identification for robust source extraction.

Findings

High convergence speed achieved through auxiliary function optimization

Effective extraction of the desired speaker in moving and noisy scenarios

Experimental analysis of x-vector properties in cross-talk conditions

Abstract

We propose a novel algorithm for adaptive blind audio source extraction. The proposed method is based on independent vector analysis and utilizes the auxiliary function optimization to achieve high convergence speed. The algorithm is partially supervised by a pilot signal related to the source of interest (SOI), which ensures that the method correctly extracts the utterance of the desired speaker. The pilot is based on the identification of a dominant speaker in the mixture using x-vectors. The properties of the x-vectors computed in the presence of cross-talk are experimentally analyzed. The proposed approach is verified in a scenario with a moving SOI, static interfering speaker, and environmental noise.