kohoutck/whisper-small-capstone

Whisper Small for Capstone

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

• Loss: 0.0001
• Wer: 0.0973

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: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 2
• optimizer: Use OptimizerNames.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: 500
• training_steps: 4000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0142	5.9172	1000	0.0121	1.8005
0.0002	11.8343	2000	0.0003	0.0973
0.0002	17.7515	3000	0.0011	0.1703
0.0001	23.6686	4000	0.0001	0.0973

Framework versions

• Transformers 4.49.0
• Pytorch 2.5.0+cu124
• Datasets 3.3.2
• Tokenizers 0.21.0