p4b/limiwhisper-small-ko-dia-gs

Whisper Small Ko(Gyungsang dialect) - p4b

This model is a fine-tuned version of openai/whisper-small on the KR Dialect Speech - gyungsang dataset. It achieves the following results on the evaluation set:

• Loss: 0.2017
• Wer: 15.9300

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: 5e-07
• train_batch_size: 96
• eval_batch_size: 64
• seed: 42
• distributed_type: multi-GPU
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 500
• training_steps: 5000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.5909	0.2	1000	0.4133	211.6022
0.3612	0.4	2000	0.2137	16.9429
0.5373	0.6	3000	0.2063	15.8379
0.2909	0.8	4000	0.2012	15.8379
0.3317	1.0	5000	0.2017	15.9300

Framework versions

• Transformers 4.26.0.dev0
• Pytorch 1.14.0.dev20221208+cu116
• Datasets 2.7.1.dev0
• Tokenizers 0.13.2