janstusio/Whisper_Large_V3_Turbo_Pl_-_Jan_Stusio

Whisper Large V3 Turbo Pl - Jan Stusio

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

• Loss: 0.2419
• Wer: 11.5953

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: 1
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 16
• optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 50
• training_steps: 500
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1101	1.592	100	0.1981	12.4077
0.0538	3.176	200	0.2147	11.7061
0.0125	4.768	300	0.2200	11.8538
0.0031	6.352	400	0.2362	11.6691
0.0011	7.944	500	0.2419	11.5953

Framework versions

• Transformers 4.57.1
• Pytorch 2.9.0+cu130
• Datasets 4.3.0
• Tokenizers 0.22.1