2026-02-24_22-57-15

This model is a fine-tuned version of openai/whisper-small.en on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2768
• Wer: 11.7841

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 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: 10000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
1.0793	0.1123	500	0.3269	12.3113
0.2697	0.2247	1000	0.3019	13.0615
0.2525	0.3370	1500	0.2933	12.5373
0.2553	0.4493	2000	0.2903	11.4227
0.2382	0.5617	2500	0.2855	11.5509
0.2397	0.6740	3000	0.2804	11.7325
0.2394	0.7863	3500	0.2768	11.7841

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1