Grammars

Grammars - rules for tokenization, classification into semiotic classes, tagging and finally verbalization into spoken form. Rules are written with Pynini and are essentially character-level WFSTs.

en_us_normalization.production

Text normalization rules for english are adapted from NVIDIA: https://github.com/NVIDIA/NeMo/tree/main/nemo_text_processing/text_normalization/en