This model is a fine-tuned version of openai/whisper-small on the aihub elder over 70 dataset. It achieves the following results on the evaluation set:

Loss: 0.1811
Cer: 5.4335

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: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
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: 50
num_epochs: 2
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer
0.3481	0.13	100	0.2725	7.0019
0.2173	0.26	200	0.2210	6.0679
0.184	0.39	300	0.2115	5.8388
0.1949	0.52	400	0.2045	5.6685
0.2008	0.64	500	0.2024	6.4850
0.1712	0.77	600	0.1937	6.1854
0.1504	0.9	700	0.1948	5.5569
0.0714	1.03	800	0.1867	5.1809
0.0706	1.16	900	0.1894	5.5216
0.0784	1.29	1000	0.1904	5.3102
0.0766	1.42	1100	0.1889	5.8094
0.0663	1.55	1200	0.1835	5.7448
0.0751	1.68	1300	0.1815	5.3219
0.0585	1.81	1400	0.1808	5.2044
0.0728	1.93	1500	0.1811	5.4335

Framework versions

Transformers 4.37.0.dev0
Pytorch 2.1.0+cu121
Datasets 2.16.0
Tokenizers 0.15.0

Gummybear05/whisper-small-ko-E30_Yfreq

whisper-small-ko-E30_Yfreq

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Gummybear05/
whisper-small-ko-E30_Yfreq