whisper-large-v3-ft-btb-ca-cy

This model is a fine-tuned version of openai/whisper-large-v3 on the DewiBrynJones/banc-trawsgrifiadau-bangor-clean train main, cymen-arfor/25awr train+dev main dataset. It achieves the following results on the evaluation set:

• Loss: 0.3810
• Wer: 0.2750

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
0.5152	0.5411	1000	0.4954	0.3535
0.3339	1.0823	2000	0.4205	0.3198
0.3189	1.6234	3000	0.3911	0.2913
0.2051	2.1645	4000	0.3863	0.2790
0.202	2.7056	5000	0.3810	0.2750

Framework versions

• Transformers 4.46.2
• Pytorch 2.5.1+cu124
• Datasets 3.1.0
• Tokenizers 0.20.3