facuvillegas/whisper-large-arg_riopl

Whisper Large Turbo Es - Facundo Villegas

This model is a fine-tuned version of openai/whisper-large-v3-turbo on the Common Voice 11.0 - ES Rio Platense dataset. It achieves the following results on the evaluation set:

• Loss: 0.3191
• Wer: 10.9474

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: 500
• training_steps: 5000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1046	1.4577	1000	0.2665	12.8947
0.0529	2.9155	2000	0.2579	12.1053
0.012	4.3732	3000	0.2881	11.1053
0.0032	5.8309	4000	0.2995	11.6842
0.0006	7.2886	5000	0.3191	10.9474

Framework versions

• Transformers 4.48.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0