whisper-medium-loz

This model is a fine-tuned version of openai/whisper-medium on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7116
• Wer: 83.0853

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: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 4
• 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
2.5661	1.08	500	2.4638	75.4238
0.7885	2.16	1000	1.5264	57.4479
0.5592	3.23	1500	1.3387	56.4385
0.3583	4.31	2000	1.1329	62.6116
0.2569	5.39	2500	0.7550	84.4700
0.1977	6.47	3000	0.6823	97.2564
0.1697	7.54	3500	0.7048	89.2584
0.1754	8.62	4000	0.7116	83.0853

Framework versions

• Transformers 4.27.0.dev0
• Pytorch 1.13.1+cu116
• Datasets 2.9.0
• Tokenizers 0.13.2