whisper-large-tr

This model is a fine-tuned version of openai/whisper-large-v3 on the arrow dataset. It achieves the following results on the evaluation set:

• Loss: 0.2053
• Wer: 12.6174

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: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 100
• training_steps: 2000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0116	2.7933	500	0.1709	13.0201
0.0007	5.5866	1000	0.1830	13.4228
0.0001	8.3799	1500	0.1998	12.6174
0.0003	11.1732	2000	0.2053	12.6174

Framework versions

• Transformers 5.4.0
• Pytorch 2.10.0+cu128
• Datasets 4.8.4
• Tokenizers 0.22.2