LG-L3

This model is a fine-tuned version of Sunbird/asr-whisper-large-v3-salt on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.7580
• Wer: 0.6216
• Cer: 0.2515

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 64
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: polynomial
• lr_scheduler_warmup_steps: 100
• training_steps: 2000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
0.4753	10.7598	1000	0.9098	0.6793	0.3620
0.3513	21.5102	2000	1.0081	0.6811	0.3490

Framework versions

• PEFT 0.18.1
• Transformers 4.52.0
• Pytorch 2.10.0+cu128
• Datasets 3.6.0
• Tokenizers 0.21.4