Speech enhancement with LSTM recurrent neural networks and its application to noise-robust ASR

Warning The system is temporarily closed to updates for reporting purpose.

Weninger, Felix and Erdoğan, Hakan and Watanabe, Shinji and Vincent, Emmanuel and Le Roux, Jonathan and Hershey, John R. and Schuller, Bjoern (2015) Speech enhancement with LSTM recurrent neural networks and its application to noise-robust ASR. In: 12th International Conference on Latent Variable Analysis and Signal Separation (LVA/ICA), Liberec

Full text not available from this repository. (Request a copy)

Abstract

We evaluate some recent developments in recurrent neural network (RNN) based speech enhancement in the light of noise-robust automatic speech recognition (ASR). The proposed framework is based on Long Short-Term Memory (LSTM) RNNs which are discriminatively trained according to an optimal speech reconstruction objective. We demonstrate that LSTM speech enhancement, even when used 'naively' as front-end processing, delivers competitive results on the CHiME-2 speech recognition task. Furthermore, simple, feature-level fusion based extensions to the framework are proposed to improve the integration with the ASR back-end. These yield a best result of 13.76% average word error rate, which is, to our knowledge, the best score to date.
Item Type: Papers in Conference Proceedings
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions: Faculty of Engineering and Natural Sciences > Academic programs > Electronics
Faculty of Engineering and Natural Sciences
Depositing User: Hakan Erdoğan
Date Deposited: 24 Dec 2015 16:06
Last Modified: 26 Apr 2022 09:21
URI: https://research.sabanciuniv.edu/id/eprint/28861

Actions (login required)

View Item
View Item