Improved MVDR Beamforming Using LSTM Speech Models To Clean Spatial Clustering Masks

Abstract

Spatial clustering techniques can achieve significant multi-channel noise reduction across relatively arbitrary microphone configurations, but have difficulty incorporating a detailed speech/noise model. In contrast, LSTM neural networks have successfully been trained to recognize speech from noise on single-channel inputs, but have difficulty taking full advantage of the information in multi-channel recordings. This paper integrates these two approaches, training LSTM speech models to clean the masks generated by the Model-based EM Source Separation and Localization (MESSL) spatial clustering method. By doing so, it attains both the spatial separation performance and generality of multi-channel spatial clustering and the signal modeling performance of multiple parallel single-channel LSTM speech enhancers. Our experiments show that when our system is applied to the CHiME-3 dataset of noisy tablet recordings, it increases speech quality as measured by the Perceptual Evaluation of Speec

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyni2020improved

Related papers

• Combining Spatial Clustering With LSTM Speech Models For Multichannel Speech Enhancement (2020)0.00
• Deep Long Short-term Memory Adaptive Beamforming Networks For Multichannel Robust Speech Recognition (2017)13.23
• Unsupervised Speech Enhancement Based On Multichannel Nmf-informed Beamforming For Noise-robust Automatic Speech Recognition (2019)13.23
• ADL-MVDR: All Deep Learning MVDR Beamformer For Target Speech Separation (2020)15.00
• Multi-talker MVDR Beamforming Based On Extended Complex Gaussian Mixture Model (2019)0.00
• Multichannel Loss Function For Supervised Speech Source Separation By Mask-based Beamforming (2019)7.50
• Multi-geometry Spatial Acoustic Modeling For Distant Speech Recognition (2019)6.34
• Student-teacher Learning For BLSTM Mask-based Speech Enhancement (2018)9.59