cdli-whisper-en-ug-ke-openai-large-v3-full-a40-1500-fresh

This model is a fine-tuned version of openai/whisper-large-v3 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6603
• Wer: 0.2533
• Cer: 0.1741

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: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• 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: constant_with_warmup
• lr_scheduler_warmup_steps: 150
• training_steps: 1500

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
0.5387	1.6967	500	0.6719	0.2632	0.1803
0.5173	3.3908	1000	0.6613	0.2599	0.1791
0.412	5.0850	1500	0.6603	0.2533	0.1741

Framework versions

• Transformers 4.57.6
• Pytorch 2.7.1+cu118
• Datasets 3.6.0
• Tokenizers 0.22.2