CaSNet: Compress-and-Send Network Based Multi-Device Speech Enhancement Model for Distributed Microphone Arrays
Chengqian Jiang, Jie Zhang, Haoyin Yan
TL;DR
CaSNet addresses bandwidth bottlenecks in distributed microphone array speech enhancement by moving processing to edge devices and transmitting compact, SVD-derived features to a reference fusion center. The FC uses cross-window query to align asynchronous, compressed cues and decodes them with a dual-path architecture to produce spatially coherent enhanced speech. The approach yields substantial data savings (NSA) with negligible degradation in standard SE metrics and scales to arbitrary numbers of microphones, matching or surpassing state-of-the-art performance on WSJ0-WHAM! and RealMAN datasets. This work enables more practical, energy-efficient DMAs and suggests avenues for adaptive bitrate control and sensor scheduling in future systems.
Abstract
Distributed microphone array (DMA) is a promising next-generation platform for speech interaction, where speech enhancement (SE) is still required to improve the speech quality in noisy cases. Existing SE methods usually first gather raw waveforms at a fusion center (FC) from all devices and then design a multi-microphone model, causing high bandwidth and energy costs. In this work, we propose a \emph{Compress-and-Send Network (CaSNet)} for resource-constrained DMAs, where one microphone serves as the FC and reference. Each of other devices encodes the measured raw data into a feature matrix, which is then compressed by singular value decomposition (SVD) to produce a more compact representation. The received features at the FC are aligned via cross window query with respect to the reference, followed by neural decoding to yield spatially coherent enhanced speech. Experiments on multiple datasets show that the proposed CaSNet can save the data amount with a negligible impact on the performance compared to the uncompressed case. The reproducible code is available at https://github.com/Jokejiangv/CaSNet.