Inter-channel Conv-TasNet for multichannel speech enhancement

TL;DR

This paper introduces an advanced multichannel speech enhancement network based on Conv-TasNet, which effectively exploits inter-channel relationships and spatial information to significantly improve speech quality and noise suppression.

Contribution

It extends Conv-TasNet into a multichannel framework that fully utilizes inter-channel relationships and spatial information, outperforming existing models with fewer parameters.

Findings

Outperforms state-of-the-art multichannel neural networks

Uses fewer parameters while achieving better enhancement

Significant improvements in SDR, PESQ, and STOI on CHiME-3

Abstract

Speech enhancement in multichannel settings has been realized by utilizing the spatial information embedded in multiple microphone signals. Moreover, deep neural networks (DNNs) have been recently advanced in this field; however, studies on the efficient multichannel network structure fully exploiting spatial information and inter-channel relationships is still in its early stages. In this study, we propose an end-to-end time-domain speech enhancement network that can facilitate the use of inter-channel relationships at individual layers of a DNN. The proposed technique is based on a fully convolutional time-domain audio separation network (Conv-TasNet), originally developed for speech separation tasks. We extend Conv-TasNet into several forms that can handle multichannel input signals and learn inter-channel relationships. To this end, we modify the encoder-mask-decoder structures of the network to be compatible with 3-D tensors defined over spatial channels, features, and time dimensions. In particular, we conduct extensive parameter analyses on the convolution structure and propose independent assignment of the depthwise and 1$\times$1 convolution layers to the feature and spatial dimensions, respectively. We demonstrate that the enriched inter-channel information from the proposed network plays a significant role in suppressing noisy signals impinging from various directions. The proposed inter-channel Conv-TasNet outperforms the state-of-the-art multichannel variants of neural networks, even with one-tenth of their parameter size. The performance of the proposed model is evaluated using the CHiME-3 dataset, which exhibits a remarkable improvement in SDR, PESQ, and STOI.