Informed Source Separation using Iterative Reconstruction

TL;DR

This paper introduces an iterative informed source separation method that enhances reconstruction quality of mixed audio signals, outperforming traditional Wiener-based techniques but with higher computational cost.

Contribution

It proposes a novel iterative reconstruction algorithm with a dual resolution approach for improved transient detail in source separation.

Findings

Outperforms Wiener-based ISS by up to 3dB in distortion

Uses iterative time-frequency consistency and re-mixing constraints

Achieves sharper transient reconstruction

Abstract

This paper presents a technique for Informed Source Separation (ISS) of a single channel mixture, based on the Multiple Input Spectrogram Inversion method. The reconstruction of the source signals is iterative, alternating between a time- frequency consistency enforcement and a re-mixing constraint. A dual resolution technique is also proposed, for sharper transients reconstruction. The two algorithms are compared to a state-of-the-art Wiener-based ISS technique, on a database of fourteen monophonic mixtures, with standard source separation objective measures. Experimental results show that the proposed algorithms outperform both this reference technique and the oracle Wiener filter by up to 3dB in distortion, at the cost of a significantly heavier computation.