ninninz/whisper-ckm-2

whisper-large-v3-croarian

This model is a fine-tuned version of openai/whisper-large-v3 on the audiofolder dataset. It achieves the following results on the evaluation set:

• Loss: 1.7755
• Wer: 64.1494

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: 16
• eval_batch_size: 8
• 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: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0169	22.73	1000	1.3990	48.7258
0.0005	45.45	2000	1.6605	56.3042
0.0002	68.18	3000	1.7494	61.4410
0.0001	90.91	4000	1.7755	64.1494

Framework versions

• Transformers 4.36.1
• Pytorch 2.1.1+cu121
• Datasets 2.15.0
• Tokenizers 0.15.0