Downloads · 30 days
0
0% of all-time downloads
conformer-ctc-small-fluencybank
arielcerdap/conformer-ctc-small-fluencybank
conformer-ctc-small-fluencybank is a automatic speech recognition model from arielcerdap. Use it when you need speech turned into text. It is set up for nemo.
Fine-tuned from nvidia/sttenconformerctcsmall on the FluencyBank Timestamped dataset (arielcerdap/TimeStamped-Splits), targeting verbatim transcription of stuttered speech from adults who stutter (PWS).
All-time downloads
32
Public
Repo size
53.3 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.nemo53.3 MB · 100%
At a glance
- Task
- Automatic Speech Recognition
- Library
- nemo
- Access
- Public
- Created
- Mar 26, 2026
- Updated
- Mar 26, 2026
- SHA
- b7b2422e
Datasets
Model card
From the Hugging Face model README
Conformer-CTC fine-tuned on FluencyBank Timestamped
Fine-tuned from nvidia/stt_en_conformer_ctc_small on the FluencyBank Timestamped dataset
(arielcerdap/TimeStamped-Splits),
targeting verbatim transcription of stuttered speech from adults who stutter (PWS).
Results on test set
| Metric | Value |
|---|---|
| WER | 18.26% |
Training details
- Base model:
nvidia/stt_en_conformer_ctc_small - Dataset:
arielcerdap/TimeStamped-Splits(train=2,744 / val=342 / test=342) - Target: verbatim transcription (disfluencies preserved)
- Optimizer: AdamW (lr=1e-5, weight_decay=1e-3)
- Scheduler: CosineAnnealing (warmup_ratio=0.05)
- Batch effective: 64
- Early stopping: patience=10 on val_wer
- Task
- Automatic Speech Recognition
- Library
- nemo
- Languages
- en
- Created
- Mar 26, 2026
- Updated
- Mar 26, 2026