whitefox123/whisper-large-ar5

Whisper large - tuned

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

• Loss: 0.1992
• Wer: 156.8649

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: 9375
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0864	1.6	1000	0.1155	165.5135
0.0291	3.2	2000	0.1192	268.0360
0.0196	4.8	3000	0.1317	217.9820
0.0024	6.4	4000	0.1583	136.1802
0.0012	8.0	5000	0.1708	136.3604
0.0004	9.6	6000	0.1841	128.7207
0.0009	11.2	7000	0.1831	169.8739
0.0003	12.8	8000	0.1885	158.7387
0.0001	14.4	9000	0.1992	156.8649

Framework versions

• Transformers 4.38.0.dev0
• Pytorch 2.1.0+cu118
• Datasets 2.17.0
• Tokenizers 0.15.2