Framewise phoneme classification with bidirectional LSTM networks

In this paper, we apply bidirectional training to a long short term memory (LSTM) network for the first time. We also present a modified, full gradient version of the LSTM learning algorithm. We discuss the significance of framewise phoneme classification to continuous speech recognition, and the va...

Full description

Saved in:
Bibliographic Details
Published inProceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005 Vol. 4; pp. 2047 - 2052 vol. 4
Main Authors Graves, A., Schmidhuber, J.
Format Conference Proceeding
LanguageEnglish
Published IEEE 2005
Subjects
Acoustic measurements
Data analysis
Electronic mail
Error correction
Hidden Markov models
Memory architecture
Neural networks
Recurrent neural networks
Speech recognition
Online AccessGet full text

Cover

More Information
Summary:In this paper, we apply bidirectional training to a long short term memory (LSTM) network for the first time. We also present a modified, full gradient version of the LSTM learning algorithm. We discuss the significance of framewise phoneme classification to continuous speech recognition, and the validity of using bidirectional networks for online causal tasks. On the TIMIT speech database, we measure the framewise phoneme classification scores of bidirectional and unidirectional variants of both LSTM and conventional recurrent neural networks (RNNs). We find that bidirectional LSTM outperforms both RNNs and unidirectional LSTM.
ISBN:0780390482
9780780390485
ISSN:2161-4393
DOI:10.1109/IJCNN.2005.1556215