Cristhian2430/whisper-large-coes-v6

Whisper Large SEIN - COES SEIN - Version 6

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: 1.3004
• Wer: 37.6838

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 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: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0001	166.6667	500	1.1083	39.1544
0.0	333.3333	1000	1.1779	37.5
0.0	500.0	1500	1.2174	37.5
0.0	666.6667	2000	1.2450	37.3162
0.0	833.3333	2500	1.2670	37.5
0.0	1000.0	3000	1.2838	37.5
0.0	1166.6667	3500	1.2955	37.6838
0.0	1333.3333	4000	1.3004	37.6838

Framework versions

• Transformers 4.48.0.dev0
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.21.0