jiwon65/whisper-small_korean-zeroth

whisper-small-korr

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: 0.3466
• Wer: 19.9610

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: 2e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.3119	0.69	100	0.3334	20.6884
0.1223	1.39	200	0.3179	21.4336
0.0757	2.08	300	0.3234	20.3158
0.0349	2.77	400	0.3329	20.8481
0.0172	3.47	500	0.3354	20.1916
0.0059	4.16	600	0.3357	19.7480
0.0057	4.85	700	0.3396	19.9965
0.0046	5.55	800	0.3417	19.7658
0.0025	6.24	900	0.3461	20.0497
0.0029	6.93	1000	0.3466	19.9610

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3