gencgeray/whisper-small-uk-another-test

Whisper Small uk - Herai Hench KI-11

This model is a fine-tuned version of openai/whisper-small on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.3516
• Wer: 32.3857
• Cer: 9.2793

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 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: 6000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
0.3167	0.4552	1000	0.4242	36.0647	10.0056
0.2786	0.9103	2000	0.3727	33.7793	9.3631
0.155	1.3655	3000	0.3582	32.0513	8.7764
0.1286	1.8207	4000	0.3527	32.7480	8.9953
0.0688	2.2758	5000	0.3513	32.1070	8.8882
0.074	2.7310	6000	0.3516	32.3857	9.2793

Framework versions

• Transformers 4.52.4
• Pytorch 2.5.1+cu124
• Datasets 3.6.0
• Tokenizers 0.21.0