Multiple-target deep learning for LSTM-RNN based speech enhancement
Authors:
- 1 January 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 136-140
Abstract
In this study, we explore long short-term memory recurrent neural networks (LSTM-RNNs) for speech enhancement. First, a regression LSTM-RNN approach for a direct mapping from the noisy to clean speech features is presented and verified to be more effective than deep neural network (DNN) based regression techniques in modeling long-term acoustic context. Then, a comprehensive comparison between the proposed direct mapping based LSTM-RNN and ideal ratio mask (IRM) based LSTM-RNNs is conducted. We observe that the direct mapping framework achieves better speech intelligibility at low signal-to-noise ratios (SNRs) while the IRM approach shows its superiority at high SNRs. Accordingly, to fully utilize this complementarity, a novel multiple-target joint learning approach is designed. The experiments under unseen noises show that the proposed framework can consistently and significantly improve the objective measures for both speech quality and intelligibility.Keywords
This publication has 19 references indexed in Scilit:
- On Training Targets for Supervised Speech SeparationIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2014
- On the Ideal Ratio Mask as the Goal of Computational Auditory Scene Analysis2014
- Ideal ratio mask estimation using deep neural networks for robust speech recognition2013
- A Fast Learning Algorithm for Deep Belief NetsNeural Computation, 2006
- On Ideal Binary Mask As the Computational Goal of Auditory Scene Analysis2006
- Multitask Learning1998
- Long Short-Term MemoryNeural Computation, 1997
- 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
- Speech enhancement using a minimum mean-square error log-spectral amplitude estimatorIEEE Transactions on Acoustics, Speech, and Signal Processing, 1985
- Speech enhancement using a minimum-mean square error short-time spectral amplitude estimatorIEEE Transactions on Acoustics, Speech, and Signal Processing, 1984