Gaffar-21/Whisper-Tiny-finetuned-id

Whisper Tiny - FineTuned - Id -

This model is a fine-tuned version of openai/whisper-tiny on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:

• Loss: 1.4286
• Wer: 59.8967

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: 32
• eval_batch_size: 8
• seed: 42
• optimizer: Use 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: 2
• training_steps: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
1.7163	0.0159	1	1.8245	86.1446
1.7097	0.0317	2	1.8245	86.1446
1.7377	0.0476	3	1.8245	86.1446
1.65	0.0635	4	1.8245	86.1446
1.9103	0.0794	5	1.7649	68.5026
1.7345	0.0952	6	1.6261	74.1824
1.5202	0.1111	7	1.5249	59.6386
1.5166	0.1270	8	1.4786	59.3804
1.5636	0.1429	9	1.4523	59.5525
1.5493	0.1587	10	1.4286	59.8967

Framework versions

• Transformers 4.46.3
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3