metadata
language:
- ko
license: apache-2.0
tags:
- hf-asr-leaderboard
- generated_from_trainer
datasets:
- test
base_model: openai/whisper-base
model-index:
- name: basetrained
results: []
basetrained
This model is a fine-tuned version of openai/whisper-base on the test dataset. It achieves the following results on the evaluation set:
- Loss: 0.0000
- Cer: 100.0
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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 3000
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Cer |
---|---|---|---|---|
0.0 | 6.8 | 1000 | 0.0000 | 6107.8498 |
0.0 | 13.61 | 2000 | 0.0000 | 100.0 |
0.0 | 20.41 | 3000 | 0.0000 | 100.0 |
Framework versions
- Transformers 4.40.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2