stillerman/fdt-disfluency-medium-41m
Disfluency deletion tagger for live speech transcripts: tags every
whitespace word KEEP / DELETE / KEEP_STRIP_COMMA / KEEP_CAPITALIZE,
then a ~15-line reconstruction turns tags into cleaned text. Deletion-only by
construction — it cannot rephrase, hallucinate, or alter names and numbers.
- Architecture: BERT L8/H512 (41M params), v1 data mix
- Val metrics: exact-match 0.9455, DELETE-F1 0.9887
- Training data: synthetic disfluency injection over conversational corpora —
see stillerman/fdt-disfluency-synthetic
onnx/model_quantized.onnx (int8) is ready for transformers.js
(device: "webgpu", dtype: "q8"); runs at ~10–50 ms per utterance
in-browser.- ⚠️ Trained partly on DailyDialog (CC BY-NC-SA): treat as research
artifact, not for commercial deployment as-is.
Trained on a DGX Spark as part of the FluencyAI digital-twin project.
