English-3

This model is a fine-tuned version of openai/whisper-small.en on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0008
• Wer: 8.1694

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-05
• train_batch_size: 8
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
No log	1.0	11	0.5281	37.8215
No log	2.0	22	0.1894	13.0106
No log	3.0	33	0.1160	36.7625
No log	4.0	44	0.0543	10.5900
No log	5.0	55	0.0576	12.8593
No log	6.0	66	0.0509	8.3207
No log	7.0	77	0.0121	7.5643
No log	8.0	88	0.0025	7.5643
No log	9.0	99	0.0010	8.1694
No log	10.0	110	0.0008	8.1694

Framework versions

• Transformers 4.30.0.dev0
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.13.3