mina

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

• Loss: 0.3404
• Wer: 0.2722

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: 2e-05
• train_batch_size: 32
• eval_batch_size: 32
• 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: 50
• training_steps: 12000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.7658	0.3040	500	0.4899	3.0318
0.4325	0.6079	1000	0.4118	1.6752
0.379	0.9119	1500	0.3764	2.0571
0.2677	1.2158	2000	0.3626	1.5884
0.2349	1.5198	2500	0.3496	0.3929
0.2304	1.8237	3000	0.3383	0.4325
0.1852	2.1277	3500	0.3402	0.2619
0.1351	2.4316	4000	0.3372	0.4407
0.1352	2.7356	4500	0.3346	0.2663
0.126	3.0395	5000	0.3404	0.2722

Framework versions

• Transformers 4.46.3
• Pytorch 2.4.0+cu121
• Datasets 3.6.0
• Tokenizers 0.20.3