rangsitra-whisper-sm-selected-rgss-0.25-beta-0.3 – AI Model by SmartWhatt | AlphaNeural AI

You can deploy this model and start earning money today!

SmartWhatt/
rangsitra-whisper-sm-selected-rgss-0.25-beta-0.3

like

0

automatic-speech-recognition

generated_from_trainer

mozilla-foundation/common_voice_17_0

openai/whisper-small

endpoints_compatible

Views

No views yet

Rangsitra Whisper

This model is a fine-tuned version of openai/whisper-small on the Common Voice 17.0 dataset. It achieves the following results on the evaluation set:

Cer: 9.1095
Wer: 31.0646
Loss: 0.2508

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-05
train_batch_size: 16
eval_batch_size: 16
seed: 42
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: warmup_stable_decay
lr_scheduler_warmup_steps: 506
num_epochs: 5
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Cer	Wer	Validation Loss
0.3936	1.0	1014	12.7788	42.3454	0.2772
0.2432	2.0	2028	10.5945	36.0968	0.2385
0.0984	3.0	3042	10.1979	33.7773	0.2415
0.0286	4.0	4056	9.3353	31.3840	0.2438
0.0113	5.0	5070	9.1095	31.0646	0.2508

Framework versions

Transformers 4.52.4
Pytorch 2.9.1+cu128
Datasets 3.6.0
Tokenizers 0.21.4