dataprizma/whisper-large-v3-turbo

Whisper Large v3 Turbo - Bahriddin Muminov

This model is a fine-tuned version of openai/whisper-large-v3-turbo on the Common Voice 16.1 dataset. It achieves the following results on the evaluation set:

• Loss: 0.2958
• Wer: 28.2582

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: 1000
• training_steps: 10000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.429	0.66	2000	0.4073	38.0018
0.2671	1.32	4000	0.3378	31.0778
0.2511	1.98	6000	0.3102	29.2484
0.1539	2.64	8000	0.3022	30.0763
0.111	3.3	10000	0.2958	28.2582

Framework versions

• Transformers 4.37.2
• Pytorch 2.2.2+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2