leenag/Norm_Malasar_Luke

This model is a fine-tuned version of openai/whisper-small on the Spoken Bible Corpus: Malasar dataset. It achieves the following results on the evaluation set:

• Loss: 0.5217
• Wer: 52.4656

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: 32
• eval_batch_size: 16
• 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: 2000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1406	11.3636	250	0.2856	55.7339
0.0084	22.7273	500	0.4196	53.8417
0.0022	34.0909	750	0.4641	53.3257
0.0005	45.4545	1000	0.4835	51.6628
0.0002	56.8182	1250	0.5049	52.0642
0.0002	68.1818	1500	0.5149	52.4656
0.0002	79.5455	1750	0.5200	52.2936
0.0002	90.9091	2000	0.5217	52.4656

Framework versions

• Transformers 4.41.2
• Pytorch 2.0.1+cu117
• Datasets 2.16.0
• Tokenizers 0.19.1