TasNet: time-domain audio separation network for real-time, single-channel speech separation

TL;DR

TasNet introduces a time-domain neural network for real-time, single-channel speech separation that outperforms frequency-domain methods, reducing latency and computational cost, suitable for low-power applications.

Contribution

The paper presents TasNet, a novel time-domain approach that directly models signals, eliminating the need for frequency decomposition and improving real-time speech separation performance.

Findings

Outperforms state-of-the-art speech separation algorithms

Reduces computational cost significantly

Minimizes latency for real-time applications

Abstract

Robust speech processing in multi-talker environments requires effective speech separation. Recent deep learning systems have made significant progress toward solving this problem, yet it remains challenging particularly in real-time, short latency applications. Most methods attempt to construct a mask for each source in time-frequency representation of the mixture signal which is not necessarily an optimal representation for speech separation. In addition, time-frequency decomposition results in inherent problems such as phase/magnitude decoupling and long time window which is required to achieve sufficient frequency resolution. We propose Time-domain Audio Separation Network (TasNet) to overcome these limitations. We directly model the signal in the time-domain using an encoder-decoder framework and perform the source separation on nonnegative encoder outputs. This method removes the frequency decomposition step and reduces the separation problem to estimation of source masks on encoder outputs which is then synthesized by the decoder. Our system outperforms the current state-of-the-art causal and noncausal speech separation algorithms, reduces the computational cost of speech separation, and significantly reduces the minimum required latency of the output. This makes TasNet suitable for applications where low-power, real-time implementation is desirable such as in hearable and telecommunication devices.