Luandrie/_Whisper_Call_Center_en_lr8_warm100

Distill Whisper Call Center Tforge Dev lr8

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

• Loss: 1.3425
• Wer: 48.4839

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-08
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• training_steps: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.8109	3.0722	1000	1.3788	49.5262
0.6552	6.1444	2000	1.3500	48.9103
0.6757	9.2166	3000	1.3437	48.2470
0.6513	12.2888	4000	1.3425	48.4839

Framework versions

• Transformers 4.45.2
• Pytorch 2.7.0+cu126
• Datasets 3.6.0
• Tokenizers 0.20.3