whisper-tiny-kor_eng_tiny_oc_op

CianKim/whisper-tiny-kor_eng_tiny_oc_op

whisper-tiny-kor_eng_tiny_oc_op is a automatic speech recognition model from CianKim. Use it when you need speech turned into text. It is set up for transformers.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

18% of all-time downloads

All-time downloads

Public

Parameters

37.8M

5.7 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors151 MB · 99%

At a glance

Task: Automatic Speech Recognition
Library: transformers
Model type: whisper
Access: Public
Created: May 8, 2025
Updated: May 15, 2025
SHA: eb4a6370

Model card

From the Hugging Face model README

whisper-tiny-kor_eng_tiny_oc_op

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.5599
Cer: 12.4240

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: 3e-05
train_batch_size: 12
eval_batch_size: 6
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
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	Cer
0.727	1.8868	100	0.5359	13.5792
0.2838	3.7736	200	0.4147	10.6607
0.0922	5.6604	300	0.4501	10.0730
0.0333	7.5472	400	0.4309	8.6948
0.024	9.4340	500	0.4911	8.1678
0.0213	11.3208	600	0.4632	9.5460
0.0156	13.2075	700	0.4925	9.0191
0.0097	15.0943	800	0.5143	7.2558
0.007	16.9811	900	0.4931	12.9307
0.0059	18.8679	1000	0.5075	10.1338
0.0029	20.7547	1100	0.5118	9.9919
0.0017	22.6415	1200	0.5100	9.7284
0.001	24.5283	1300	0.5111	10.3567
0.0012	26.4151	1400	0.5236	11.4309
0.0008	28.3019	1500	0.5289	9.0799
0.0001	30.1887	1600	0.5309	8.8569
0.0001	32.0755	1700	0.5337	8.9785
0.0001	33.9623	1800	0.5357	8.7556
0.0001	35.8491	1900	0.5380	8.7556
0.0001	37.7358	2000	0.5400	8.7353
0.0001	39.6226	2100	0.5419	8.7353
0.0001	41.5094	2200	0.5444	8.7353
0.0001	43.3962	2300	0.5458	9.3636
0.0001	45.2830	2400	0.5473	9.4041
0.0001	47.1698	2500	0.5487	9.4447
0.0001	49.0566	2600	0.5502	9.4041
0.0001	50.9434	2700	0.5512	9.4447
0.0001	52.8302	2800	0.5527	8.6745
0.0	54.7170	2900	0.5537	8.7150
0.0	56.6038	3000	0.5546	8.7150
0.0	58.4906	3100	0.5558	12.4443
0.0	60.3774	3200	0.5564	12.4443
0.0	62.2642	3300	0.5572	12.4240
0.0	64.1509	3400	0.5579	12.4443
0.0	66.0377	3500	0.5584	12.4240
0.0	67.9245	3600	0.5590	12.4037
0.0	69.8113	3700	0.5592	12.4240
0.0	71.6981	3800	0.5595	12.4240
0.0	73.5849	3900	0.5597	12.4240
0.0	75.4717	4000	0.5599	12.4240

Framework versions

Transformers 4.51.3
Pytorch 2.6.0+cu126
Datasets 3.5.0
Tokenizers 0.21.1

Task: Automatic Speech Recognition
Library: transformers
Type: whisper
Created: May 8, 2025
Updated: May 15, 2025