Oyounghyun/contents

study0703

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

• Loss: 0.1988
• Cer: 6.6313

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• 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	Cer
0.004	9.6154	1000	0.1731	6.3102
0.0007	19.2308	2000	0.1849	6.3584
0.0004	28.8462	3000	0.1921	6.4226
0.0003	38.4615	4000	0.1968	6.5992
0.0002	48.0769	5000	0.1988	6.6313

Framework versions

• Transformers 4.43.0.dev0
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1