iulik-pisik/vreme_model_base

Whisper Base Romanian - Vreme

This model is a fine-tuned version of openai/whisper-base on the Vreme ProTv dataset. It achieves the following results on the evaluation set:

• Loss: 0.2275
• Wer: 11.5701

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
0.0284	6.02	1000	0.1553	12.2483
0.0013	12.05	2000	0.2104	11.8783
0.0006	18.07	3000	0.2232	11.5290
0.0004	24.1	4000	0.2275	11.5701

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2