Bridging the Gap Between Monaural Speech Enhancement and Recognition With Distortion-Independent Acoustic Modeling
- 17 October 2019
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Audio, Speech, and Language Processing
- Vol. 28 (23299290) , 39-48
Abstract
Monaural speech enhancement has made dramatic advances since the introduction of deep learning a few years ago. Although enhanced speech has been demonstrated to have better intelligibility and quality for human listeners, feeding it directly to automatic speech recognition (ASR) systems trained with noisy speech has not produced expected improvements in ASR performance. The lack of an enhancement benefit on recognition, or the gap between monaural speech enhancement and recognition, is often attributed to speech distortions introduced in the enhancement process. In this article, we analyze the distortion problem, compare different acoustic models, and investigate a distortion-independent training scheme for monaural speech recognition. Experimental results suggest that distortion-independent acoustic modeling is able to overcome the distortion problem. Such an acoustic model can also work with speech enhancement models different from the one used during training. Moreover, the models investigated in this paper outperform the previous best system on the CHiME-2 corpus.Keywords
Funding Information
- National Science Foundation (IIS-1409431, ECCS-1808932)
- Ohio Supercomputer Center
This publication has 25 references indexed in Scilit:
- A Joint Training Framework for Robust Automatic Speech RecognitionIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2016
- Phase-sensitive and recognition-boosted speech separation using deep recurrent neural networks2015
- Joint training of front-end and back-end deep neural networks for robust speech recognition2015
- A Regression Approach to Speech Enhancement Based on Deep Neural NetworksIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2014
- On Training Targets for Supervised Speech SeparationIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2014
- Deep stacking networks with time series for speech separation2014
- An algorithm to improve speech recognition in noise for hearing-impaired listenersThe Journal of the Acoustical Society of America, 2013
- Ideal ratio mask estimation using deep neural networks for robust speech recognition2013
- The second ‘chime’ speech separation and recognition challenge: Datasets, tasks and baselines2013
- Speech enhancement using a minimum-mean square error short-time spectral amplitude estimatorIEEE Transactions on Acoustics, Speech, and Signal Processing, 1984