Norm_KLuke_Med / README.md
leenag's picture
End of training
21b654d verified
metadata
language:
  - ymr
license: apache-2.0
base_model: openai/whisper-small
tags:
  - generated_from_trainer
metrics:
  - wer
model-index:
  - name: leenag/Norm_KLuke_Med
    results: []

leenag/Norm_KLuke_Med

This model is a fine-tuned version of openai/whisper-small on the Spoken Bible Corpus: Malasar dataset. It achieves the following results on the evaluation set:

  • Loss: 0.5144
  • Wer: 44.9541

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: 32
  • 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: 2000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
0.0275 11.3636 250 0.3935 48.5092
0.0091 22.7273 500 0.4626 49.0252
0.0012 34.0909 750 0.4798 47.1330
0.0002 45.4545 1000 0.5015 46.3876
0.0001 56.8182 1250 0.4947 45.5849
0.0 68.1818 1500 0.5088 45.0688
0.0 79.5455 1750 0.5129 44.9541
0.0 90.9091 2000 0.5144 44.9541

Framework versions

  • Transformers 4.41.2
  • Pytorch 2.0.1+cu117
  • Datasets 2.16.0
  • Tokenizers 0.19.1