Model
Russian single-pass tagger-based model for inverse text normalization based on BERT encoder, trained on 2 mln sentences from Google Text Normalization Dataset, achieves 3.55% WER on Google default test set
Use the NGC CLI to download:
Copied!
| Name | Size | Updated | Actions |
|---|---|---|---|
itn_ru_thutmose_bert.nemo | 695.47 MB | February 28, 2025 UTC |