EUROSPEECH 2003 - INTERSPEECH 2003
We present a novel data-driven language independent approach for grapheme to phoneme conversion, which achieves a phoneme error rate of 3.68% and a pronunciation error rate of 17.13% for English. We apply our stochastic model to the task of dictionary verification and conclude that it is able to detect spurious entries, which can then be examined and corrected by a human expert.
Bibliographic reference. Vozila, Paul / Adams, Jeff / Lobacheva, Yuliya / Thomas, Ryan (2003): "Grapheme to phoneme conversion and dictionary verification using graphonemes", In EUROSPEECH-2003, 2469-2472.