Subjective and Objective Quality Assessment of Audio Source Separation

Authors:
V. Emiya;E. Vincent;N. Harlander;V. Hohmann
Affiliations:
Centre Inria Rennes-Bretagne Atlantique, INRIA, Rennes, France;-;-;-
Venue:
IEEE Transactions on Audio, Speech, and Language Processing
Year:
2011

Citing 0
Cited 8

The 2010 signal separation evaluation campaign (SiSEC2010): audio source separation

LVA/ICA'10 Proceedings of the 9th international conference on Latent variable analysis and signal separation
The signal separation evaluation campaign (2007-2010): Achievements and remaining challenges

Signal Processing
Informed source separation through spectrogram coding and data embedding

Signal Processing
The 2011 signal separation evaluation campaign (SiSEC2011): - audio source separation -

LVA/ICA'12 Proceedings of the 10th international conference on Latent Variable Analysis and Signal Separation
Improved perceptual metrics for the evaluation of audio source separation

LVA/ICA'12 Proceedings of the 10th international conference on Latent Variable Analysis and Signal Separation
Model-Driven speech enhancement for multisource reverberant environment (signal separation evaluation campaign (SiSEC) 2011)

LVA/ICA'12 Proceedings of the 10th international conference on Latent Variable Analysis and Signal Separation
Perceptually enhanced blind single-channel music source separation by Non-negative Matrix Factorization

Digital Signal Processing
Noise variance estimation based on dual-channel phase difference for speech enhancement

Digital Signal Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

We aim to assess the perceived quality of estimated source signals in the context of audio source separation. These signals may involve one or more kinds of distortions, including distortion of the target source, interference from the other sources or musical noise artifacts. We propose a subjective test protocol to assess the perceived quality with respect to each kind of distortion and collect the scores of 20 subjects over 80 sounds. We then propose a family of objective measures aiming to predict these subjective scores based on the decomposition of the estimation error into several distortion components and on the use of the PEMO-Q perceptual salience measure to provide multiple features that are then combined. These measures increase correlation with subjective scores up to 0.5 compared to nonlinear mapping of individual state-of-the-art source separation measures. Finally, we released the data and code presented in this paper in a freely available toolkit called PEASS.