whisper-small-ko-EYfreq_speed – AI Model by Gummybear05 | AlphaNeural AI

You can deploy this model and start earning money today!

Gummybear05/
whisper-small-ko-EYfreq_speed

like

0

automatic-speech-recognition

hf-asr-leaderboard

generated_from_trainer

openai/whisper-small

endpoints_compatible

Views

No views yet

whisper-small-ko-EYfreq_speed

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

Loss: 0.2194
Cer: 5.0258

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: 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: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Cer
0.3544	0.3	100	0.2530	6.6160
0.31	0.59	200	0.2367	6.4024
0.2798	0.89	300	0.2177	5.8862
0.1122	1.19	400	0.2188	7.2153
0.1128	1.48	500	0.2272	5.9396
0.0975	1.78	600	0.2157	5.6963
0.0326	2.07	700	0.2100	5.3166
0.0355	2.37	800	0.2149	5.1326
0.0401	2.67	900	0.2180	5.9752
0.0352	2.96	1000	0.2083	5.0614
0.0147	3.26	1100	0.2134	5.1386
0.0126	3.56	1200	0.2141	5.0377
0.0119	3.85	1300	0.2213	4.9783
0.005	4.15	1400	0.2164	4.9071
0.0043	4.44	1500	0.2155	4.9309
0.0042	4.74	1600	0.2194	5.0258

Framework versions

Transformers 4.34.0.dev0
Pytorch 2.0.1+cu118
Datasets 2.14.5
Tokenizers 0.13.3