On Training Targets for Supervised Speech Separation
- 28 August 2014
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Audio, Speech, and Language Processing
- Vol. 22 (12), 1849-1858
- https://doi.org/10.1109/taslp.2014.2352935
Abstract
Formulation of speech separation as a supervised learning problem has shown considerable promise. In its simplest form, a supervised learning algorithm, typically a deep neural network, is trained to learn a mapping from noisy features to a time-frequency representation of the target of interest. Traditionally, the ideal binary mask (IBM) is used as the target because of its simplicity and large speech intelligibility gains. The supervised learning framework, however, is not restricted to the use of binary targets. In this study, we evaluate and compare separation results by using different training targets, including the IBM, the target binary mask, the ideal ratio mask (IRM), the short-time Fourier transform spectral magnitude and its corresponding mask (FFT-MASK), and the Gammatone frequency power spectrum. Our results in various test conditions reveal that the two ratio mask targets, the IRM and the FFT-MASK, outperform the other targets in terms of objective intelligibility and quality metrics. In addition, we find that masking based targets, in general, are significantly better than spectral envelope based targets. We also present comparisons with recent methods in non-negative matrix factorization and speech enhancement, which show clear performance advantages of supervised speech separation.This publication has 32 references indexed in Scilit:
- The role of binary mask patterns in automatic speech recognition in background noiseThe Journal of the Acoustical Society of America, 2013
- A classification based approach to speech segregationThe Journal of the Acoustical Society of America, 2012
- Super-human multi-talker speech recognition: A graphical modeling approachComputer Speech & Language, 2010
- Role of mask pattern in intelligibility of ideal binary-masked noisy speechThe Journal of the Acoustical Society of America, 2009
- An algorithm that improves speech intelligibility in noise for normal-hearing listenersThe Journal of the Acoustical Society of America, 2009
- On the optimality of ideal binary time–frequency masksSpeech Communication, 2009
- Factors influencing intelligibility of ideal binary-masked speech: Implications for noise reductionThe Journal of the Acoustical Society of America, 2008
- Binary and ratio time-frequency masks for robust speech recognitionSpeech Communication, 2006
- Determination of the Potential Benefit of Time-Frequency Gain ManipulationEar & Hearing, 2006
- Assessment for automatic speech recognition: II. NOISEX-92: A database and an experiment to study the effect of additive noise on speech recognition systemsSpeech Communication, 1993