Modelpublic

dianavdavidson/wh_l_v3_cv_common_voices_52393__val_finetune_trial

sourceHugging Faceapache-2.0updated 6mo agoView on Hugging Face

0likes13downloads

Model Card

whlv3cvcommonvoices52393_valfinetune_trial

This model is a fine-tuned version of openai/whisper-large-v3 on the commonvoice22_0 dataset. It achieves the following results on the evaluation set:

—Loss: 0.1578
—Global Wer: 37.5974

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: 5e-06
—trainbatchsize: 4
—evalbatchsize: 4
—seed: 42
—optimizer: Use adamwtorchfused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
—lrschedulertype: constantwithwarmup
—lrschedulerwarmup_steps: 50
—training_steps: 500

Training results

Training Loss	Epoch	Step	Validation Loss	Global Wer
0.3931	0.0411	50	0.3158	27.6440
0.2087	0.0821	100	0.2110	27.3479
0.1857	0.1232	150	0.1808	36.8964
0.1680	0.1642	200	0.1784	38.2375
0.1803	0.2053	250	0.1657	45.7482
0.1338	0.2463	300	0.1617	40.6191
0.1510	0.2874	350	0.1646	36.2999
0.1456	0.3284	400	0.1519	42.9660
0.1440	0.3695	450	0.1529	42.3913
0.1410	0.4105	500	0.1578	37.5974

Framework versions

—Transformers 5.0.0.dev0
—Pytorch 2.9.0+cu126
—Datasets 3.6.0
—Tokenizers 0.22.2