Kwimp/no_augmentation

Whisper Small

This model is a fine-tuned version of openai/whisper-small on the Speechocean762_CMUkids_Myst dataset. It achieves the following results on the evaluation set:

• Loss: 0.4056
• Wer: 18.0759

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: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• 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: 2048
• training_steps: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
4.9694	1.8398	500	3.0150	18.1672
3.1539	3.6777	1000	1.8826	16.9380
1.7307	5.5157	1500	1.0186	17.3722
0.7948	7.3536	2000	0.4598	18.7217
0.6236	9.1915	2500	0.4215	19.6081
0.6026	11.0295	3000	0.4111	17.6885
0.5911	12.8692	3500	0.4069	18.4567
0.6070	14.7072	4000	0.4056	18.0759

Framework versions

• Transformers 5.8.1
• Pytorch 2.5.1+cu121
• Datasets 4.8.5
• Tokenizers 0.22.2