Real-time Streaming Wave-U-Net with Temporal Convolutions for Multichannel Speech Enhancement

by   Vasiliy Kuzmin, et al.

In this paper, we describe the work that we have done to participate in Task1 of the ConferencingSpeech2021 challenge. This task set a goal to develop the solution for multi-channel speech enhancement in a real-time manner. We propose a novel system for streaming speech enhancement. We employ Wave-U-Net architecture with temporal convolutions in encoder and decoder. We incorporate self-attention in the decoder to apply attention mask retrieved from skip-connection on features from down-blocks. We explore history cache mechanisms that work like hidden states in recurrent networks and implemented them in proposal solution. It helps us to run an inference with chunks length 40ms and Real-Time Factor 0.4 with the same precision.


page 1

page 2

page 3

page 4


Dense CNN with Self-Attention for Time-Domain Speech Enhancement

Speech enhancement in the time domain is becoming increasingly popular i...

Improved Speech Enhancement with the Wave-U-Net

We study the use of the Wave-U-Net architecture for speech enhancement, ...

Towards end-to-end speech enhancement with a variational U-Net architecture

In this paper, we investigate the viability of a variational U-Net archi...

Iterative autoregression: a novel trick to improve your low-latency speech enhancement model

Streaming models are an essential component of real-time speech enhancem...

Efficient Encoder-Decoder and Dual-Path Conformer for Comprehensive Feature Learning in Speech Enhancement

Current speech enhancement (SE) research has largely neglected channel a...

Cross-Attention is all you need: Real-Time Streaming Transformers for Personalised Speech Enhancement

Personalised speech enhancement (PSE), which extracts only the speech of...

Please sign up or login with your details

Forgot password? Click here to reset