metadata
library_name: transformers
license: apache-2.0
base_model: openai/whisper-large-v3
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: whisper-large-v3-ft-btbn-ca
results: []
whisper-large-v3-ft-btbn-ca
This model is a fine-tuned version of openai/whisper-large-v3 on the DewiBrynJones/banc-trawsgrifiadau-bangor train main, cymen-arfor/15awr train+dev main dataset. It achieves the following results on the evaluation set:
- Loss: 0.4127
- Wer: 0.2775
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: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- 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_steps: 500
- training_steps: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.9964 | 0.6555 | 1000 | 0.4889 | 0.3663 |
| 0.6606 | 1.3110 | 2000 | 0.4223 | 0.3117 |
| 0.6065 | 1.9666 | 3000 | 0.3859 | 0.2873 |
| 0.3894 | 2.6221 | 4000 | 0.3962 | 0.2787 |
| 0.2478 | 3.2776 | 5000 | 0.4127 | 0.2775 |
Framework versions
- Transformers 4.46.1
- Pytorch 2.5.1+cu124
- Datasets 3.1.0
- Tokenizers 0.20.1