Downloads · 30 days
16
25% of all-time downloads
fdt-disfluency-distilbert-66m
stillerman/fdt-disfluency-distilbert-66m
fdt-disfluency-distilbert-66m is a token classification model from stillerman. Use it when you need labels on individual words, such as names. It is set up for transformers.js. The card lists the license as apache-2.0.
Disfluency deletion tagger for live speech transcripts: tags every whitespace word KEEP / DELETE / KEEPSTRIPCOMMA / KEEPCAPITALIZE, then a ~15-line reconstruction turns tags into cleaned text. Deletion-only by constru…
All-time downloads
64
Public
Parameters
65.2M
326 MB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors261 MB · 80%
At a glance
- Task
- Token Classification
- Library
- transformers.js
- License
- apache-2.0
- Model type
- distilbert
- Access
- Public
- Created
- Jul 7, 2026
- Updated
- Jul 7, 2026
- SHA
- dacec369
Try a prompt
Model card
From the Hugging Face model README
stillerman/fdt-disfluency-distilbert-66m
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: distilbert-base-cased (65.8M params), v1 data mix
- Val metrics: exact-match 0.9760, DELETE-F1 0.9935
- 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.
- Task
- Token Classification
- Library
- transformers.js
- Type
- distilbert
- License
- apache-2.0
- Created
- Jul 7, 2026
- Updated
- Jul 7, 2026