Kwimp/pitch_speed_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: 0.4057
• Wer: 18.0848

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
5.2459	1.6734	500	3.0174	17.7129
3.2713	3.3451	1000	1.8842	17.5214
1.7679	5.0168	1500	1.0173	17.6885
0.8126	6.6901	2000	0.4594	18.5570
0.6499	8.3618	2500	0.4213	19.0870
0.6065	10.0335	3000	0.4108	18.5881
0.5906	11.7069	3500	0.4070	17.6083
0.5759	13.3786	4000	0.4057	18.0848

Framework versions

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