alpcansoydas/whisper-large-v2-tr-ft-15-03-26-partial-encoder-only

Whisper for Turkish Call Centers

This model is a fine-tuned version of openai/whisper-large-v2 on the Custom turkish call center simulated data dataset. It achieves the following results on the evaluation set:

• Loss: 0.2925

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-06
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• 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: 4500
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.3092	0.2222	250	0.3472
0.3042	0.4444	500	0.3181
0.2641	0.6667	750	0.3088
0.2671	0.8889	1000	0.3017
0.2298	1.1111	1250	0.2988
0.2118	1.3333	1500	0.2972
0.2090	1.5556	1750	0.2944
0.2341	1.7778	2000	0.2919
0.2238	2.0	2250	0.2896
0.2318	2.2222	2500	0.2921
0.1918	2.4444	2750	0.2918
0.2273	2.6667	3000	0.2914
0.2058	2.8889	3250	0.2906
0.2249	3.1111	3500	0.2921
0.1645	3.3333	3750	0.2931
0.1897	3.5556	4000	0.2924
0.2128	3.7778	4250	0.2925
0.2373	4.0	4500	0.2925

Framework versions

• Transformers 5.3.0
• Pytorch 2.10.0+cu128
• Datasets 4.7.0
• Tokenizers 0.22.2