whisper-large-v3-turbo

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

• Loss: 0.2416
• Wer: 0.0960

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: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.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: 500
• training_steps: 5000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1951	1.4820	1000	0.2532	0.1386
0.1032	2.9640	2000	0.2232	0.1189
0.0242	4.4449	3000	0.2313	0.1090
0.0142	5.9270	4000	0.2315	0.1016
0.0018	7.4079	5000	0.2416	0.0960

Framework versions

• Transformers 4.57.0.dev0
• Pytorch 2.8.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.0