whisper-tiny-lin-x2athanase

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

• Loss: 0.7646
• Wer: 0.3957
• Cer: 0.1800

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: 0.0001
• train_batch_size: 32
• eval_batch_size: 64
• 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: 500
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
0.5878	1.2255	500	0.7013	0.4613	0.2346
0.4241	2.4510	1000	0.6461	0.4066	0.1939
0.3457	3.6765	1500	0.6477	0.4275	0.2055
0.2568	4.9020	2000	0.7025	0.3885	0.1820
0.1118	6.1275	2500	0.7646	0.3957	0.1800

Framework versions

• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2