Fine-tune 資訊

• 原始模型: openai/whisper-large-v3
• 使用音訊數量: 17696
• 使用音訊總長: 9.42 小時
• 音訊平均長度: 1.92 秒
• GPU: NVIDIA H100 PCIe x 1
• 訓練時間: 02:08:15
• 模型大小: 5.75 GB
• 訓練參數:
  • batch size: 80
  • eval batch size: 40
  • gradient checkpointing: True
  • fp16: False
  • bf16: True

Fine-tuned Whisper model for Legislative Yuan of Taiwan

• Loss: 0.0235
• Wer: 78.2926

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

• learning_rate: 1e-05
• train_batch_size: 80
• eval_batch_size: 40
• 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: 50
• training_steps: 500

Training results

Framework versions

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