Unified Language-Independent DNN-Based G2P Converter

Markéta Jůzová, Daniel Tihelka, Jakub Vít


We introduce a unified Grapheme-to-phoneme conversion framework based on the composition of deep neural networks. In contrary to the usual approaches building the G2P frameworks from the dictionary, we use whole phrases, which allows us to capture various language properties, e.g. cross-word assimilation, without the need for any special care or topology adjustments. The evaluation is carried out on three different languages — English, Czech and Russian. Each requires dealing with specific properties, stressing the proposed framework in various ways. The very first results show promising performance of the proposed framework, dealing with all the phenomena specific to the tested languages. Thus, we consider the framework to be language-independent for a wide range of languages.


 DOI: 10.21437/Interspeech.2019-2335

Cite as: Jůzová, M., Tihelka, D., Vít, J. (2019) Unified Language-Independent DNN-Based G2P Converter. Proc. Interspeech 2019, 2085-2089, DOI: 10.21437/Interspeech.2019-2335.


@inproceedings{Jůzová2019,
  author={Markéta Jůzová and Daniel Tihelka and Jakub Vít},
  title={{Unified Language-Independent DNN-Based G2P Converter}},
  year=2019,
  booktitle={Proc. Interspeech 2019},
  pages={2085--2089},
  doi={10.21437/Interspeech.2019-2335},
  url={http://dx.doi.org/10.21437/Interspeech.2019-2335}
}