The Applications Of Unsupervised Learning To Japanese Grapheme-Phoneme Alignment

In this paper, we adapt the TF-IDF model to the Japanese grapheme-phoneme alignment task, by way of a simple statistical model and an incremental learning method. In the incremental learning method, grapheme-phoneme alignment paradigms are disambiguated one at a t ime according to the relative plausibility of the highest scoring alignment schema, and the statistical model is re-trained accordingly. On limited evaluation, the learning method achieved an accuracy of 93.28%, representing a slight improvement over a baseline rule-based method.