whisper-a-nomimose-ls

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

• Loss: 0.0225
• Wer: 39.3068

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: 0.0004
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Use 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: 132
• num_epochs: 11
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.9697	1.0	109	0.1804	43.4366
0.1049	2.0	218	0.0635	24.7050
0.0649	3.0	327	0.0376	13.4218
0.0659	4.0	436	0.0545	12.0944
0.0301	5.0	545	0.0538	23.2301
0.0382	6.0	654	0.0335	25.0737
0.0176	7.0	763	0.0253	28.9086
0.0153	8.0	872	0.0258	26.9174
0.0082	9.0	981	0.0257	51.4749
0.0054	10.0	1090	0.0222	42.5516
0.0036	10.9032	1188	0.0225	39.3068

Framework versions

• Transformers 4.47.0.dev0
• Pytorch 2.4.0
• Datasets 3.0.1
• Tokenizers 0.20.0