Low bit rate binaural link for improved ultra low-latency low-complexity multichannel speech enhancement in Hearing Aids

TL;DR

This paper introduces GCFSnet, a low-latency deep learning model for speech enhancement in hearing aids, utilizing a binaural link and quantization to operate efficiently within hardware constraints.

Contribution

The paper presents GCFSnet, a novel causal multichannel speech enhancement model optimized for hearing aids with low latency and computational efficiency, incorporating a low bit rate binaural link.

Findings

GCFSnet matches oracle binaural beamformer performance on objective metrics.

The model operates with a latency of only 2ms.

Quantization-aware training reduces model size and complexity.

Abstract

Speech enhancement in hearing aids is a challenging task since the hardware limits the number of possible operations and the latency needs to be in the range of only a few milliseconds. We propose a deep-learning model compatible with these limitations, which we refer to as Group-Communication Filter-and-Sum Network (GCFSnet). GCFSnet is a causal multiple-input single output enhancement model using filter-and-sum processing in the time-frequency domain and a multi-frame deep post filter. All filters are complex-valued and are estimated by a deep-learning model using weight-sharing through Group Communication and quantization-aware training for reducing model size and computational footprint. For a further increase in performance, a low bit rate binaural link for delayed binaural features is proposed to use binaural information while retaining a latency of 2ms. The performance of an oracle binaural LCMV beamformer in non-low-latency configuration can be matched even by a unilateral configuration of the GCFSnet in terms of objective metrics.