DianaJin/Dianajin

jinkrsmodel

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

• Loss: 5.7864
• Cer: 100.0

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: 400
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer
2.7143	50.0	100	4.6669	294.9367
0.0018	100.0	200	5.2253	100.0
0.0008	150.0	300	5.4916	100.0
0.0005	200.0	400	5.7864	100.0

Framework versions

• Transformers 4.36.0.dev0
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0