stillerman/fdt-disfluency-small-29m-v2

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 L4/H512 (28.5M params), v2 data mix (substitution repairs, collisions, doubled fillers)
• Val metrics: exact-match 0.9255, DELETE-F1 0.9874
• 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.