Hugging Face's logo

lsuarezgo96
/
whisper-large-v3-ar

PEFT
TensorBoard
Safetensors
Arabic
Generated from Trainer
Model card Files Metrics Community
whisper-large-v3-ar / README.md
lsuarezgo96's picture
lsuarezgo96
End of training
c4c7750 verified
preview code
|
raw
history blame contribute delete
1.59 kB
metadata
base_model: openai/whisper-large-v3
datasets:
  - khaledalganem/sada2022
language:
  - ar
library_name: peft
license: apache-2.0
tags:
  - generated_from_trainer
model-index:
  - name: Whisper Large v3 Ar-luis.s
    results: []

Whisper Large v3 Ar-luis.s

This model is a fine-tuned version of openai/whisper-large-v3 on the SADA 2022 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.6510

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: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 50
  • num_epochs: 3
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
0.4047 1.0 105 0.6539
0.5493 2.0 210 0.6151
0.1803 3.0 315 0.6510

Framework versions

  • PEFT 0.12.1.dev0
  • Transformers 4.45.0.dev0
  • Pytorch 2.4.0+cu121
  • Datasets 2.19.2
  • Tokenizers 0.19.1