Grapheme-to-Phoneme Transformer Model for Transfer Learning Dialects
This work addresses the challenge of building robust G2P models for dialects with small pronunciation dictionaries, which is incremental as it adapts existing transformer methods to improve generalization in low-data scenarios.
The paper tackles the problem of grapheme-to-phoneme conversion for dialects with limited data by proposing a transformer-based model that uses transfer learning, achieving a phoneme error rate of 2.469% on a test set after fine-tuning, compared to 26.877% when trained from scratch.
Grapheme-to-Phoneme (G2P) models convert words to their phonetic pronunciations. Classic G2P methods include rule-based systems and pronunciation dictionaries, while modern G2P systems incorporate learning, such as, LSTM and Transformer-based attention models. Usually, dictionary-based methods require significant manual effort to build, and have limited adaptivity on unseen words. And transformer-based models require significant training data, and do not generalize well, especially for dialects with limited data. We propose a novel use of transformer-based attention model that can adapt to unseen dialects of English language, while using a small dictionary. We show that our method has potential applications for accent transfer for text-to-speech, and for building robust G2P models for dialects with limited pronunciation dictionary size. We experiment with two English dialects: Indian and British. A model trained from scratch using 1000 words from British English dictionary, with 14211 words held out, leads to phoneme error rate (PER) of 26.877%, on a test set generated using the full dictionary. The same model pretrained on CMUDict American English dictionary, and fine-tuned on the same dataset leads to PER of 2.469% on the test set.