volkan-aslan/whisper-tiny-tr-optimized-v3

Whisper Tiny TR Optimized V3 - Volkan ASLAN

This model is a fine-tuned version of openai/whisper-tiny on the Common Voice 17.0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.4976
• Wer: 38.8691

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: 3.75e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• 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
• training_steps: 5000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.4323	0.6894	1000	0.5809	44.9016
0.2368	1.3785	2000	0.5329	41.3199
0.1354	2.0676	3000	0.5079	41.0066
0.1287	2.7570	4000	0.4931	38.8487
0.0888	3.4461	5000	0.4976	38.8691

Framework versions

• Transformers 4.51.3
• Pytorch 2.5.1+cu121
• Datasets 3.5.1
• Tokenizers 0.21.1