Focus on the Sound around You: Monaural Target Speaker Extraction via Distance and Speaker Information

TL;DR

This paper introduces a monaural target speaker extraction method that utilizes distance and speaker information, achieving robust performance in noisy, reverberant environments without prior speaker enrollment.

Contribution

The paper proposes the NS-Extractor leveraging distance info and self-enrollment, with full- and sub-band modeling to improve robustness in reverberant conditions.

Findings

Effective in cross-dataset evaluations

Outperforms existing methods in noisy environments

Robust to reverberation effects

Abstract

Previously, Target Speaker Extraction (TSE) has yielded outstanding performance in certain application scenarios for speech enhancement and source separation. However, obtaining auxiliary speaker-related information is still challenging in noisy environments with significant reverberation. inspired by the recently proposed distance-based sound separation, we propose the near sound (NS) extractor, which leverages distance information for TSE to reliably extract speaker information without requiring previous speaker enrolment, called speaker embedding self-enrollment (SESE). Full- & sub-band modeling is introduced to enhance our NS-Extractor's adaptability towards environments with significant reverberation. Experimental results on several cross-datasets demonstrate the effectiveness of our improvements and the excellent performance of our proposed NS-Extractor in different application scenarios.