CasNet: Investigating Channel Robustness for Speech Separation

TL;DR

CasNet is a novel deep learning framework that enhances speech separation robustness by incorporating channel embeddings, effectively addressing channel mismatch issues in real-world scenarios.

Contribution

Introduces CasNet, a channel-aware speech separation network that leverages channel embeddings and FiLM to improve performance under channel mismatch conditions.

Findings

CasNet outperforms baseline TasNet in experiments.

Channel embeddings improve robustness to channel mismatch.

Training strategies influence the role of channel information.

Abstract

Recording channel mismatch between training and testing conditions has been shown to be a serious problem for speech separation. This situation greatly reduces the separation performance, and cannot meet the requirement of daily use. In this study, inheriting the use of our previously constructed TAT-2mix corpus, we address the channel mismatch problem by proposing a channel-aware audio separation network (CasNet), a deep learning framework for end-to-end time-domain speech separation. CasNet is implemented on top of TasNet. Channel embedding (characterizing channel information in a mixture of multiple utterances) generated by Channel Encoder is introduced into the separation module by the FiLM technique. Through two training strategies, we explore two roles that channel embedding may play: 1) a real-life noise disturbance, making the model more robust, or 2) a guide, instructing the separation model to retain the desired channel information. Experimental results on TAT-2mix show that CasNet trained with both training strategies outperforms the TasNet baseline, which does not use channel embeddings.