whisper-large-v3-khmer-5hr – AI Model by Tnaot | AlphaNeural AI

You can deploy this model and start earning money today!

Tnaot/
whisper-large-v3-khmer-5hr

like

0

automatic-speech-recognition

generated_from_trainer

unsloth/whisper-large-v3

endpoints_compatible

Views

No views yet

whisper-large-v3-khmer-5hr

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

Loss: 0.1697
Ter: 8770.2703
Wer: 87.7703

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: 4e-05
train_batch_size: 4
eval_batch_size: 4
seed: 3407
gradient_accumulation_steps: 4
total_train_batch_size: 16
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
training_steps: 2000

Training results

Training Loss	Epoch	Step	Validation Loss	Ter	Wer
1.0361	0.0893	250	0.5756	10668.9189	106.6892
0.3132	0.1786	500	0.2967	9979.7297	99.7973
0.2511	0.2679	750	0.2337	8925.6757	89.2568
0.195	0.3571	1000	0.1950	9033.7838	90.3378
0.1781	0.4464	1250	0.1697	8770.2703	87.7703

Framework versions

Transformers 4.56.2
Pytorch 2.8.0+cu128
Datasets 4.3.0
Tokenizers 0.22.1