bagasshw/whisper-tiny-javanese-openslr-v2

Whisper Tiny Java

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

• Loss: 0.5638
• Wer: 0.4824

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: 64
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 256
• 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_ratio: 0.1
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.6676	0.8643	500	0.5638	0.4824
0.4792	1.7277	1000	0.4284	0.5330
0.3988	2.5912	1500	0.3772	0.5687
0.3565	3.4546	2000	0.3528	0.6204
0.3386	4.3181	2500	0.3406	0.6603

Framework versions

• Transformers 4.50.0.dev0
• Pytorch 2.6.0+cu126
• Datasets 3.3.2
• Tokenizers 0.21.0