TheRains/yt-special-batch4-lr6-small

Whisper Small Indonesian

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

• Loss: 0.8639
• Wer: 54.7046

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-06
• train_batch_size: 4
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 5000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
1.1374	0.09	1000	0.9854	64.9634
0.8775	0.17	2000	0.9139	66.4613
0.9735	0.26	3000	0.8845	58.6668
0.8359	0.34	4000	0.8696	59.5876
0.9089	0.43	5000	0.8639	54.7046

Framework versions

• Transformers 4.31.0.dev0
• Pytorch 2.0.1+cu117
• Datasets 2.13.1
• Tokenizers 0.13.3