Cristhian2430/whisper-large-coes-v9

Whisper Large SEIN - COES SEIN - Version 9

This model is a fine-tuned version of openai/whisper-large-v3-turbo on the SEIN COES dataset. It achieves the following results on the evaluation set:

• Loss: 3.8246
• Wer: 67.8191

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: 8000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0123	90.9091	1000	3.1299	68.9362
0.0054	181.8182	2000	3.3876	69.0957
0.0002	272.7273	3000	3.5715	69.5213
0.0001	363.6364	4000	3.6546	68.3511
0.0002	454.5455	5000	3.7197	67.6596
0.0001	545.4545	6000	3.7702	67.5532
0.0001	636.3636	7000	3.8096	67.7660
0.0001	727.2727	8000	3.8246	67.8191

Framework versions

• Transformers 4.56.0.dev0
• Pytorch 2.6.0+cu124
• Datasets 4.0.0
• Tokenizers 0.21.4