This paper proposes an approach for speech recognition using recurrent neural networks (RNN) and corresponding training schema, in which RNN is trained as a digit classifier. In order to reduce the training cost and improve the performance, in our system, the teacher signals are obtained from hidden Markov model and the initial feedback values of RNN are trained along with the weights. Experiment result shows that these methods improve the recognition accuracy by %.
Trentin, E., Bengio, Y., Furlanello, C., DE MORI, R. (1998). Neural networks for speech recognition. In Spoken Dialogues with Computers (pp. 311-361). LONDON : Academic Press.
Neural networks for speech recognition
TRENTIN E.;
1998-01-01
Abstract
This paper proposes an approach for speech recognition using recurrent neural networks (RNN) and corresponding training schema, in which RNN is trained as a digit classifier. In order to reduce the training cost and improve the performance, in our system, the teacher signals are obtained from hidden Markov model and the initial feedback values of RNN are trained along with the weights. Experiment result shows that these methods improve the recognition accuracy by %.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11365/15437
Attenzione
Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo