Surt Small v1 — Training Checkpoints
This repo contains the best training checkpoint for the Surt Small v1 Gurbani ASR model.
Current Checkpoint
| Parameter | Value |
|---|---|
| Step | 3400 / 5000 |
| WER | 14.88% |
| CER | 4.30% |
| Epoch | ~3.4 |
This is the best-performing checkpoint from the full training run. The final model (step 5000, WER 15.39%) is available at surindersinghssj/surt-small-v1.
Usage
from transformers import WhisperProcessor, WhisperForConditionalGeneration
processor = WhisperProcessor.from_pretrained("openai/whisper-small")
model = WhisperForConditionalGeneration.from_pretrained("surindersinghssj/surt-small-v1-training")
For full details on training, dataset, and evaluation, see the main model card.
- Downloads last month
- 342