openai/whisper-large-v2
This model is a fine-tuned version of openai/whisper-large-v2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.8091
- Wer: 17.7875
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 32
- eval_batch_size: 16
- seed: 42
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 10000
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.2528 | 0.2 | 2000 | 0.9370 | 22.1311 |
| 0.2718 | 0.4 | 4000 | 0.8721 | 24.9294 |
| 0.2745 | 0.6 | 6000 | 0.8770 | 20.5292 |
| 0.2157 | 0.8 | 8000 | 0.8774 | 18.1018 |
| 0.1729 | 1.0 | 10000 | 0.8091 | 17.7875 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu117
- Datasets 2.7.1.dev0
- Tokenizers 0.13.2
- Downloads last month
- 71
Inference Providers
NEW
This model is not currently available via any of the supported third-party Inference Providers, and
the model is not deployed on the HF Inference API.
Spaces using arbml/whisper-largev2-ar 2
Evaluation results
- Wer on mozilla-foundation/common_voice_11_0test set self-reported38.230
- Wer on google/fleurstest set self-reported11.600