metadata
language:
- ko
license: apache-2.0
base_model: openai/whisper-base
tags:
- hf-asr-leaderboard
- generated_from_trainer
datasets:
- INo0121/low_quality_call_voice
model-index:
- name: Whisper Base for Korean Low quaiity Call Voices
results: []
Whisper Base for Korean Low quaiity Call Voices
This model is a fine-tuned version of openai/whisper-base on the Korean Low Quaiity Call Voices dataset. It achieves the following results on the evaluation set:
- Loss: 0.5102
- Cer: 31.4525
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: 8000
Training results
Training Loss | Epoch | Step | Validation Loss | Cer |
---|---|---|---|---|
0.6644 | 0.48 | 1000 | 0.6600 | 42.3757 |
0.5769 | 0.97 | 2000 | 0.5747 | 35.0448 |
0.4652 | 1.45 | 3000 | 0.5467 | 32.5385 |
0.4504 | 1.93 | 4000 | 0.5234 | 32.5923 |
0.352 | 2.42 | 5000 | 0.5199 | 33.8982 |
0.3305 | 2.9 | 6000 | 0.5097 | 31.1140 |
0.2886 | 3.38 | 7000 | 0.5124 | 32.0912 |
0.2983 | 3.86 | 8000 | 0.5102 | 31.4525 |
Framework versions
- Transformers 4.34.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3