raiyan007/whisper-tiny-6e-5

Whisper tiny bn - Raiyan

This model is a fine-tuned version of openai/whisper-tiny on the Common Voice_13.0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.1734
• Wer: 44.3491

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: 6e-05
• train_batch_size: 24
• eval_batch_size: 12
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 50
• training_steps: 3000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.261	1.0661	500	0.2417	63.3469
0.1926	2.1322	1000	0.1941	54.3987
0.1367	3.1983	1500	0.1729	49.3116
0.0994	4.2644	2000	0.1622	46.2280
0.0564	5.3305	2500	0.1669	45.0802
0.0394	6.3966	3000	0.1734	44.3491

Framework versions

• Transformers 4.41.0
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1