Hugging Face's logo

hyk000
/
foreign_p

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Korean
whisper
hf-asr-leaderboard
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
foreign_p / README.md
hyk000's picture
hyk000
End of training
11ef20b verified
preview code
|
raw
history blame
1.82 kB
metadata
library_name: transformers
language:
  - ko
license: apache-2.0
base_model: openai/whisper-base
tags:
  - hf-asr-leaderboard
  - generated_from_trainer
datasets:
  - hyk000/woerae
model-index:
  - name: wr_md
    results: []

wr_md

This model is a fine-tuned version of openai/whisper-base on the wr_ds dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0903
  • Cer: 9.2605

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: 8
  • 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: linear
  • lr_scheduler_warmup_steps: 500
  • training_steps: 4000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Cer
0.0486 3.8462 1000 0.1159 7.9222
0.0034 7.6923 2000 0.0927 11.2435
0.0015 11.5385 3000 0.0906 10.8821
0.0012 15.3846 4000 0.0903 9.2605

Framework versions

  • Transformers 4.47.0.dev0
  • Pytorch 2.5.0
  • Datasets 3.0.2
  • Tokenizers 0.20.1