whisper-turbo-fo-100h-8k-steps

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

• Loss: 0.0638
• Wer: 4.9169

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: 800
• training_steps: 8000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.2843	0.2438	1000	0.2948	21.5976
0.1829	0.4877	2000	0.1854	14.3235
0.1112	0.7315	3000	0.1511	11.5918
0.1323	0.9754	4000	0.1185	9.4021
0.0602	1.2192	5000	0.1017	7.7587
0.0416	1.4631	6000	0.0870	6.9128
0.0399	1.7069	7000	0.0717	5.4853
0.0318	1.9507	8000	0.0638	4.9169

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0