whisper-test-1

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

• Loss: 1.4129
• Cer: 29.7003

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: 48
• eval_batch_size: 24
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer
0.2285	8.33	500	0.6509	25.5488
0.0148	16.67	1000	0.7623	28.2401
0.0022	25.0	1500	0.8702	26.6749
0.0004	33.33	2000	1.0062	28.1924
0.0	41.67	2500	1.1586	28.5551
0.0	50.0	3000	1.2983	29.0704
0.0	58.33	3500	1.3890	29.3758
0.0	66.67	4000	1.4129	29.7003

Framework versions

• Transformers 4.39.1
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2