You can deploy this model and start earning money today!

alpcansoydas/
whisper-large-v2-tr-ft-01-04-26-encoder-only-50ksamples-simulated-data-lr1e-3

Name: whisper-large-v2-tr-ft-01-04-26-encoder-only-50ksamples-simulated-data-lr1e-3
Availability: InStock

automatic-speech-recognition

generated_from_trainer

openai/whisper-large-v2

finetune

apache-2.0

endpoints_compatible

Views

No views yet

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: 0.3096

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: 0.001
train_batch_size: 32
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: 2200
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.8834	0.1777	250	0.5262
0.6519	0.3554	500	0.5977
0.5224	0.5330	750	0.5000
0.4408	0.7107	1000	0.4296
0.4262	0.8884	1250	0.4085
0.3432	1.0661	1500	0.3640
0.3238	1.2438	1750	0.3395
0.3362	1.4215	2000	0.3203
0.2977	1.5636	2200	0.3096

Framework versions

Transformers 5.4.0
Pytorch 2.10.0+cu128
Datasets 4.8.4
Tokenizers 0.22.2

alpcansoydas/whisper-large-v2-tr-ft-01-04-26-encoder-only-50ksamples-simulated-data-lr1e-3

Whisper for Turkish Call Centers

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

alpcansoydas/
whisper-large-v2-tr-ft-01-04-26-encoder-only-50ksamples-simulated-data-lr1e-3