End-to-End Multi-Channel Speech Separation
This work addresses speech separation in noisy environments, offering a data-driven approach that enhances separation quality for applications like hearing aids or voice assistants, though it builds incrementally on existing end-to-end techniques.
The paper tackled multi-channel speech separation by proposing an end-to-end model that integrates waveform processing and learnable spatial features, achieving significant performance improvements on the WSJ0 far-field task compared to previous methods.
The end-to-end approach for single-channel speech separation has been studied recently and shown promising results. This paper extended the previous approach and proposed a new end-to-end model for multi-channel speech separation. The primary contributions of this work include 1) an integrated waveform-in waveform-out separation system in a single neural network architecture. 2) We reformulate the traditional short time Fourier transform (STFT) and inter-channel phase difference (IPD) as a function of time-domain convolution with a special kernel. 3) We further relaxed those fixed kernels to be learnable, so that the entire architecture becomes purely data-driven and can be trained from end-to-end. We demonstrate on the WSJ0 far-field speech separation task that, with the benefit of learnable spatial features, our proposed end-to-end multi-channel model significantly improved the performance of previous end-to-end single-channel method and traditional multi-channel methods.