phucd/whisper-small-ko

Whisper - Korean Model

This model is a fine-tuned version of openai/whisper-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4784
• Wer: 98.8922

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: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0002	52.63	1000	0.4055	107.5529
0.0001	105.26	2000	0.4449	102.2155
0.0001	157.89	3000	0.4692	93.1521
0.0001	210.53	4000	0.4784	98.8922

Framework versions

• Transformers 4.36.0.dev0
• Pytorch 1.12.1
• Datasets 2.15.0
• Tokenizers 0.15.0