luyotw/test-batch14-small-XieLongJie-11-36

Fine-tune 資訊

• 原始模型: openai/whisper-small
• 使用音訊數量: 5044
• 使用音訊總長: 2.87 小時
• 音訊平均長度: 2.05 秒
• GPU: NVIDIA H100 PCIe x 1
• 訓練時間: 00:25:30
• 模型大小: 0.90 GB
• 訓練參數:
  • batch size: 14
  • eval batch size: 7
  • gradient checkpointing: False
  • fp16: False
  • bf16: True

---

Fine-tuned Whisper model for Legislative Yuan of Taiwan

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

• Loss: 0.0416
• Wer: 88.6288

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: 14
• eval_batch_size: 7
• seed: 42
• optimizer: Use 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: 5000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0169	2.7701	1000	0.0314	87.5585
0.0028	5.5402	2000	0.0365	89.0970
0.0003	8.3102	3000	0.0396	88.2274
0.0002	11.0803	4000	0.0410	88.2943
0.0001	13.8504	5000	0.0416	88.6288

Framework versions

• Transformers 4.51.3
• Pytorch 2.5.1
• Datasets 3.5.0
• Tokenizers 0.21.1