whisper-tiny-th-4

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: 0.7699
• Cer: 31.3508

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5.0

Training results

Training Loss	Epoch	Step	Validation Loss	Cer
0.9187	1.0	38	0.8350	35.2823
0.5349	2.0	76	0.7711	31.5524
0.3241	3.0	114	0.7630	31.9556
0.2342	4.0	152	0.7662	36.0887
0.2037	5.0	190	0.7699	31.3508

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu121
• Datasets 3.0.0
• Tokenizers 0.19.1