whisper-small-chamorro

This model is a fine-tuned version of openai/whisper-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4949
• Wer: 68.9189

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: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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
0.0002	50.0	500	0.4287	74.3243
0.0001	100.0	1000	0.4461	64.8649
0.0000	150.0	1500	0.4552	64.8649
0.0000	200.0	2000	0.4682	66.2162
0.0000	250.0	2500	0.4775	68.9189
0.0000	300.0	3000	0.4872	68.9189
0.0000	350.0	3500	0.4928	68.9189
0.0000	400.0	4000	0.4949	68.9189

Framework versions

• Transformers 5.18.0
• Pytorch 2.11.0+cu130
• Datasets 5.1.0
• Tokenizers 0.23.2