David-Mazi/whisper-tiny.en-vox

Whisper Tiny En Vox - David Mazi

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

• Loss: 0.0019
• Wer: 70.0633

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: 16
• eval_batch_size: 8
• seed: 42
• 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_steps: 50
• training_steps: 500

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.2694	4.0	100	0.1224	77.4886
0.0232	8.0	200	0.0165	69.1336
0.0035	12.0	300	0.0032	80.3548
0.0022	16.0	400	0.0021	68.5044
0.0019	20.0	500	0.0019	70.0633

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1
• Datasets 3.2.0
• Tokenizers 0.21.0