patipol-bkk/whisper-tiny-alphadigit-decoder-finetune-10k-clean

Whisper Tiny AlphaDigit Decoder Finetune 10k Clean

This model is a fine-tuned version of openai/whisper-tiny on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0615
• Wer: 58.8931

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: 32
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 10000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0987	0.51	1000	0.0878	90.7605
0.0745	1.03	2000	0.0737	91.3320
0.0647	1.54	3000	0.0689	83.4997
0.063	2.05	4000	0.0676	79.2696
0.0498	2.56	5000	0.0655	69.8125
0.0511	3.08	6000	0.0634	69.5636
0.0373	3.59	7000	0.0623	65.4287
0.0361	4.1	8000	0.0619	61.4768
0.0465	4.61	9000	0.0617	59.5491
0.0376	5.13	10000	0.0615	58.8931

Framework versions

• Transformers 4.33.0.dev0
• Pytorch 2.0.1+cu117
• Datasets 2.14.4
• Tokenizers 0.13.3