Performance measurement in blind audio source separation

TLDR

In blind audio source separation, acceptable distortions between estimated and true sources vary by application. The paper evaluates blind audio source separation algorithms. Four distortion sets—from time‑invariant gains to time‑varying filters—are examined; the estimated source is decomposed into true source plus error terms for interferences, noise, and artifacts, and a global energy‑ratio measure and separate error‑term measures are derived. The measures are computed and discussed on several BASS problems of varying difficulty.

Abstract

In this paper, we discuss the evaluation of blind audio source separation (BASS) algorithms. Depending on the exact application, different distortions can be allowed between an estimated source and the wanted true source. We consider four different sets of such allowed distortions, from time-invariant gains to time-varying filters. In each case, we decompose the estimated source into a true source part plus error terms corresponding to interferences, additive noise, and algorithmic artifacts. Then, we derive a global performance measure using an energy ratio, plus a separate performance measure for each error term. These measures are computed and discussed on the results of several BASS problems with various difficulty levels