octava/whisper-small-id

ID Optimized Whisper Small for Inspirasi

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

• Loss: 0.2100
• Wer: 8.7207

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: 16
• eval_batch_size: 8
• seed: 42
• 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: 500
• training_steps: 5000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0106	5.3763	1000	0.1857	10.2083
0.0009	10.7527	2000	0.1943	9.0740
0.0004	16.1290	3000	0.2017	8.7951
0.0002	21.5054	4000	0.2077	8.7579
0.0002	26.8817	5000	0.2100	8.7207

Framework versions

• Transformers 4.51.0
• Pytorch 2.2.0a0+81ea7a4
• Datasets 3.5.1
• Tokenizers 0.21.1