DCCRN-KWS: An Audio Bias Based Model for Noise Robust Small-Footprint Keyword Spotting

Real-world complex acoustic environments especially the ones with a low signal-to-noise ratio (SNR) will bring tremendous challenges to a keyword spotting (KWS) system. Inspired by the recent advances of neural speech enhancement and context bias in speech recognition, we propose a robust audio context bias based DCCRN-KWS model to address this challenge. We form the whole architecture as a multi-task learning framework for both denoising and keyword spotting, where the DCCRN encoder is connected with the KWS model. Helped with the denoising task, we further introduce an audio context bias module to leverage the real keyword samples and bias the network to better discriminate keywords in noisy conditions. Feature merge and complex context linear modules are also introduced to strengthen such discrimination and to effectively leverage contextual information respectively. Experiments on an internal challenging dataset and the HIMIYA public dataset show that DCCRN-KWS is superior in performance, while the ablation study demonstrates the good design of the whole model.