Customizing Grapheme-to-Phoneme System for Non-Trivial Transcription Problems in
Proceedings of the 2019 Conference of the North
Grapheme to phoneme (G2P) conversion is an integral part in various text and speech processing systems, such as: Text to Speech system, Speech Recognition system, etc. The existing methodologies for G2P conversion in Bangla language are mostly rule-based. However, data-driven approaches have proved their superiority over rule-based approaches for largescale G2P conversion in other languages, such as: English, German, etc. As the performance of data-driven approaches for G2P conversion depend
... onversion depend largely on pronunciation lexicon on which the system is trained, in this paper, we investigate on developing an improved training lexicon by identifying and categorizing the critical cases in Bangla language and include those critical cases in training lexicon for developing a robust G2P conversion system in Bangla language. Additionally, we have incorporated nasal vowels in our proposed phoneme list. Our methodology outperforms other stateof-the-art approaches for G2P conversion in Bangla language. . 2016. Polyglot neural language models: A case study in cross-lingual phonetic representation learning. arXiv preprint arXiv: 1605 .03832. Macherey, et al. 2016. Google's neural machine translation system: Bridging the gap between human and machine translation. arXiv preprint arXiv:1609.08144. Kaisheng Yao and Geoffrey Zweig. 2015. Sequenceto-sequence neural net models for graphemeto-phoneme conversion. arXiv preprint arXiv:1506.00196.