RadioSES: mmWave-Based Audioradio Speech Enhancement and Separation System

TL;DR

RadioSES is a novel audioradio system that uses millimeter-wave sensing and deep learning to enhance and separate speech in noisy environments, outperforming audio-only and audiovisual methods in various settings.

Contribution

It introduces a new audioradio framework combining radio sensing with audio processing for improved speech separation and enhancement.

Findings

RadioSES outperforms state-of-the-art baselines in diverse environments.

Achieves approximately 3 dB gains in SiSDR over other methods.

Offers lower computational complexity and better privacy compared to audiovisual systems.

Abstract

Speech enhancement and separation have been a long-standing problem, especially with the recent advances using a single microphone. Although microphones perform well in constrained settings, their performance for speech separation decreases in noisy conditions. In this work, we propose RadioSES, an audioradio speech enhancement and separation system that overcomes inherent problems in audio-only systems. By fusing a complementary radio modality, RadioSES can estimate the number of speakers, solve source association problem, separate and enhance noisy mixture speeches, and improve both intelligibility and perceptual quality. We perform millimeter-wave sensing to detect and localize speakers, and introduce an audioradio deep learning framework to fuse the separate radio features with the mixed audio features. Extensive experiments using commercial off-the-shelf devices show that RadioSES outperforms a variety of state-of-the-art baselines, with consistent performance gains in different environmental settings. Compared with the audiovisual methods, RadioSES provides similar improvements (e.g., ~3 dB gains in SiSDR), along with the benefits of lower computational complexity and being less privacy concerning.