Bitwise Source Separation on Hashed Spectra: An Efficient Posterior Estimation Scheme Using Partial Rank Order Metrics

TL;DR

This paper introduces a novel, efficient bitwise source separation method using hashed spectra and partial rank order metrics, enabling fast, iteration-free posterior estimation for single-channel speech denoising.

Contribution

It presents the first iteration-free dictionary-based source separation algorithm leveraging hashed spectra and partial rank order metrics for efficient posterior probability computation.

Findings

Achieves 6-8 dB mean SDR in speech denoising

Provides a fast, iteration-free separation process

Demonstrates robustness with partial rank order features

Abstract

This paper proposes an efficient bitwise solution to the single-channel source separation task. Most dictionary-based source separation algorithms rely on iterative update rules during the run time, which becomes computationally costly especially when we employ an overcomplete dictionary and sparse encoding that tend to give better separation results. To avoid such cost we propose a bitwise scheme on hashed spectra that leads to an efficient posterior probability calculation. For each source, the algorithm uses a partial rank order metric to extract robust features that form a binarized dictionary of hashed spectra. Then, for a mixture spectrum, its hash code is compared with each source's hashed dictionary in one pass. This simple voting-based dictionary search allows a fast and iteration-free estimation of ratio masking at each bin of a signal spectrogram. We verify that the proposed BitWise Source Separation (BWSS) algorithm produces sensible source separation results for the single-channel speech denoising task, with 6-8 dB mean SDR. To our knowledge, this is the first dictionary based algorithm for this task that is completely iteration-free in both training and testing.