Joint-sequence models for grapheme-to-phoneme conversion
Speech Communication
A joint decoding algorithm for multiple-example-based addition of words to a pronunciation lexicon
ICASSP '09 Proceedings of the 2009 IEEE International Conference on Acoustics, Speech and Signal Processing
Hi-index | 0.00 |
This paper describes work in progress towards using non-phonemic respellings as an additional source of information besides spelling in the process of extending pronunciation lexicons for speech recognition and text-to-speech systems. Preliminary experimental data indicates that the approach is likely to be successful. The major benefit of the approach is that it makes extending pronunciation lexicons accessible to average users.