Litcius/Paper detail

Real-M: Towards Speech Separation on Real Mixtures

Cem Subakan, Mirco Ravanelli, Samuele Cornell, François Grondin

2022ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)17 citationsDOI

Abstract

In recent years, deep learning based source separation has achieved impressive results. Most studies, however, still evaluate separation models on synthetic datasets, while the performance of state-of-the-art techniques on in-the-wild speech data remains an open question. This paper contributes to fill this gap in two ways. First, we release the REAL-M dataset, a crowd-sourced corpus of real-life mixtures. Secondly, we address the problem of performance evaluation of real-life mixtures, where the ground truth is not available. We bypass this issue by carefully designing a blind Scale-Invariant Signal-to-Noise Ratio (SI-SNR) neural estimator. Through a user study, we show that our estimator reliably evaluates the separation performance on real mixtures, i.e. we observe that the performance predictions of the SI-SNR estimator correlate well with human opinions. Moreover, when evaluating popular speech separation models, we observe that the performance trends predicted by our estimator on the REAL-M dataset closely follow the performance trends achieved on synthetic benchmarks.

Topics & Concepts

Computer scienceEstimatorSeparation (statistics)Ground truthArtificial intelligenceDeep neural networksSource separationSpeech recognitionMachine learningDeep learningMathematicsStatisticsSpeech and Audio ProcessingMusic and Audio ProcessingSpeech Recognition and Synthesis