Whisper-Small-TF-TIMIT-FLEUR-Normalizado

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: 0.7395
• Wer: 85.3796

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 3000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.5923	1.27	500	0.9379	98.7612
0.1823	2.54	1000	0.6721	89.3262
0.0852	3.81	1500	0.6534	86.1141
0.0327	5.08	2000	0.6794	84.4019
0.0106	6.35	2500	0.7170	82.5587
0.0064	7.61	3000	0.7395	85.3796

Framework versions

• Transformers 4.28.0.dev0
• Pytorch 1.13.0
• Datasets 2.1.0
• Tokenizers 0.13.2