Configuration Parsing
Warning:
In adapter_config.json: "peft.task_type" must be a string
Whisper Medium
This model is a fine-tuned version of openai/whisper-medium on the b-brave-clean dataset. It achieves the following results on the evaluation set:
- Loss: 0.8245
- Wer: 69.1977
- Cer: 45.5477
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: 3e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.4
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 4.2121 | 1.0 | 251 | 4.2340 | 150.7163 | 87.2603 |
| 1.0726 | 2.0 | 502 | 1.1107 | 79.9427 | 52.9288 |
| 0.8421 | 3.0 | 753 | 0.9359 | 203.4384 | 152.2721 |
| 0.6134 | 4.0 | 1004 | 0.8391 | 102.2923 | 77.7515 |
| 0.4891 | 5.0 | 1255 | 0.8245 | 69.1977 | 45.5477 |
Framework versions
- PEFT 0.14.0
- Transformers 4.48.3
- Pytorch 2.2.0
- Datasets 3.2.0
- Tokenizers 0.21.0
- Downloads last month
- 18
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API:
The model has no pipeline_tag.
Model tree for miosipof/asr2_medium_v0.1
Base model
openai/whisper-medium