DCCRN-KWS: An Audio Bias Based Model For Noise Robust Small-footprint Keyword Spotting
2023 · Shubo Lv, Xiong Wang, Sining Sun, et al.
Abstract
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 denosing 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 iscriminate keywords in noisy conditions. Feature merge and complex context linear modules are also introduced to strength such discrimination and to effectively leverage contextual information respectively. Experiments on the internal challenging dataset and the HIMIYA public dataset show that our DCCRN-KWS system is superior
Authors
(none)
Tags
Stats
Related papers
- A Monaural Speech Enhancement Method For Robust Small-footprint Keyword Spotting (2019)0.00
- Small-footprint Keyword Spotting Using Deep Neural Network And Connectionist Temporal Classifier (2017)0.00
- Contrastive Learning With Audio Discrimination For Customizable Keyword Spotting In Continuous Speech (2024)0.00
- Phoneme-level Contrastive Learning For User-defined Keyword Spotting With Flexible Enrollment (2024)6.34
- Multi-task Network For Noise-robust Keyword Spotting And Speaker Verification Using Ctc-based Soft VAD And Global Query Attention (2020)9.41
- Streaming Small-footprint Keyword Spotting Using Sequence-to-sequence Models (2017)12.40
- Sequence Discriminative Training For Deep Learning Based Acoustic Keyword Spotting (2018)8.35
- End-to-end Keyword Spotting Using Neural Architecture Search And Quantization (2021)8.60