mmarron14/whisper-medium-cv17-es-500-steps_proc3-def3

Whisper Medium CV17 Es 500 steps with same training configuration as 500-steps_proc2-def3 -with filtering 30sec, customised optimizer, and same training_args -; but with processing of text using method 3 - María Marrón

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

• Loss: 0.1708
• Wer Ortho: 10.2104
• Wer: 5.9175

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: 2
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 32
• total_train_batch_size: 64
• 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
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer Ortho	Wer
0.2126	0.2	100	0.1915	11.4950	7.0113
0.1861	0.4	200	0.1843	11.2696	6.5645
0.1902	0.6	300	0.1785	10.6012	6.2633
0.1741	0.8	400	0.1735	10.3084	5.9830
0.1796	1.0	500	0.1708	10.2104	5.9175

Framework versions

• Transformers 4.53.2
• Pytorch 2.9.1+cu128
• Datasets 2.14.4
• Tokenizers 0.21.4