YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

scribe-punct-ru

Russian punctuation + capitalization model for Scribe offline voice dictation. Duplicated from ekhodzitsky/rupunct-small-onnx (MIT) so Scribe does not depend on a third-party repo; plus a Scribe sidecar.

Files

File Purpose
rupunct_small_int8.onnx (~29 MB, original) Quantized model, CPU ~1 ms per phrase
tokenizer.json (original) Fast tokenizer
model.json (Scribe sidecar) Label map (UPPER/LOWER/TOTAL + mark) and decoding style
README.md This card

Label scheme

33 classes: case of this word (UPPER/LOWER/TOTAL) + mark after it (PERIOD/COMMA/QUESTION/TIRE/VOSKL/DVOETOCHIE/PERIODCOMMA/DEFIS/QUESTIONVOSKL/MNOGOTOCHIE/O).

Usage in Scribe

Downloaded by Scribe's model manager into models/punct/ru/, runs fully offline via ONNX Runtime. No torch/transformers needed at runtime.

License / attribution

MIT. Original model by ekhodzitsky (see source repo), converted from RUPunct/RUPunct_small. Scribe sidecar (model.json, this card) is MIT as well.

Downloads last month
12
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support