whisper-a-nomi-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: 0.0301
• Wer: 97.3190

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
No log	1.0	88	0.2173	32.7078
0.9256	2.0	176	0.0358	13.6729
0.1103	3.0	264	0.1075	45.6658
0.0399	4.0	352	0.0939	68.9008
0.0319	5.0	440	0.0241	19.1242
0.0144	6.0	528	0.0252	18.2306
0.01	7.0	616	0.0300	99.3744
0.0023	8.0	704	0.0268	97.9446
0.0023	9.0	792	0.0301	97.3190
0.0003	10.0	880	0.0301	97.3190
0.0	11.0	968	0.0301	97.3190

Framework versions

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