Multichannel Speech Enhancement without Beamforming

TL;DR

This paper introduces a two-stage multichannel speech enhancement method that achieves high performance without relying on traditional beamforming, by combining a novel neural network with a recurrent network.

Contribution

It proposes a novel attentive dense convolutional network for spectrogram estimation and demonstrates that a two-stage approach with a stronger first model outperforms traditional beamforming methods.

Findings

State-of-the-art results with ADCN in single-stage models

Two-stage approach improves speech enhancement performance

Effective without traditional beamforming techniques

Abstract

Deep neural networks are often coupled with traditional spatial filters, such as MVDR beamformers for effectively exploiting spatial information. Even though single-stage end-to-end supervised models can obtain impressive enhancement, combining them with a traditional beamformer and a DNN-based post-filter in a multistage processing provides additional improvements. In this work, we propose a two-stage strategy for multi-channel speech enhancement that does not require a traditional beamformer for additional performance. First, we propose a novel attentive dense convolutional network (ADCN) for estimating real and imaginary parts of complex spectrogram. ADCN obtains state-of-the-art results among single-stage models. Next, we use ADCN with a recently proposed triple-path attentive recurrent network (TPARN) for estimating waveform samples. The proposed strategy uses two insights; first, using different approaches in two stages; and second, using a stronger model in the first stage. We illustrate the efficacy of our strategy by evaluating multiple models in a two-stage approach with and without a traditional beamformer.