Edit model card

Whisper Small Fa - BuzzyBuzzy

This model is a fine-tuned version of openai/whisper-small on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.5493
  • Wer: 34.7578

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • training_steps: 10000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
0.0179 0.86 2000 0.4881 37.4523
0.0136 1.72 4000 0.5062 37.2511
0.0037 2.58 6000 0.5282 35.6625
0.0016 3.44 8000 0.5429 35.3503
0.0003 4.3 10000 0.5493 34.7578

Framework versions

  • Transformers 4.26.0.dev0
  • Pytorch 1.13.1+cu117
  • Datasets 2.8.0
  • Tokenizers 0.13.2
Downloads last month
10
Hosted inference API
or or
This model can be loaded on the Inference API on-demand.

Dataset used to train taesiri/whisper-small-fa-3

Evaluation results