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

Loss: 0.2272
Wer: 12.4026

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: 6.25e-06
train_batch_size: 64
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: 780
training_steps: 7875
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.2238	0.2	1575	0.2392	12.4351
0.2123	0.4	3150	0.2303	12.5325
0.1962	1.0286	4725	0.2216	12.3701
0.1004	1.2286	6300	0.2309	12.2727
0.1144	1.4286	7875	0.2272	12.4026

Framework versions

Transformers 4.51.3
Pytorch 2.6.0+cu124
Datasets 3.5.0
Tokenizers 0.21.1

xbilek25/whisper-medium-en-cv-1.6

Whisper medium v1.7 - 400h

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

xbilek25/
whisper-medium-en-cv-1.6