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

Loss: 1.1564
Wer: 35.3644

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: 3e-05
train_batch_size: 48
eval_batch_size: 4
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: 210
training_steps: 2100
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
No log	0	0	2.4185	46.5401
0.8149	0.1429	300	1.0591	38.1506
0.2115	1.1429	600	1.0779	40.8757
0.0598	2.1429	900	1.1087	36.4666
0.0216	3.1429	1200	1.1280	35.9155
0.0089	4.1429	1500	1.1617	35.1806
0.0024	5.1429	1800	1.1517	34.9357
0.0012	6.1429	2100	1.1564	35.3644

Framework versions

Transformers 4.51.3
Pytorch 2.6.0+cu124
Datasets 3.5.1
Tokenizers 0.21.1

xbilek25/whisper-medium-en-cv-6.1

whisper-medium-en-cv-6.1

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

xbilek25/
whisper-medium-en-cv-6.1