EUSIPCO'2002 banner

Paper data
Application of the Neural Networks for Text-to-Phoneme Mapping

Bilcu Eniko Beatrice, Tampere University of Technology
Salmela Petri, Tampere University of Technology
Suontausta Janne, Nokia Research Center
Saarinen Jukka, Tampere University of Technology

Page numbers in the proceedings:
Volume III pp 97-100

Multimedia Data Protection / Speech Analysis and Recognition

Paper abstract
In this paper we present the results on the use of neural networks for text-to-phoneme mapping. For this mapping, we have compared the performances of the Context Dependent Multilayer Perceptron network with the Recurrent Neural Network. The results (number of parameters vs neural network model vs phoneme accuracy) are given for American English. Also some guidelines for selecting the appropriate network structure together with some methods for improving the phoneme accuracy are presented.

A PDF version is available here