Edit model card
Configuration Parsing Warning: In adapter_config.json: "peft.task_type" must be a string

Visualize in Weights & Biases

SungBeom-whisper-small-ko-multiple-bg-v1

This model is a fine-tuned version of SungBeom/whisper-small-ko on the audiofolder dataset. It achieves the following results on the evaluation set:

  • Loss: 0.7993

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-06
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.01
  • training_steps: 5000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
6.4591 17.2414 500 0.8053
6.0259 34.4828 1000 0.8035
5.7577 51.7241 1500 0.8015
5.5694 68.9655 2000 0.7999
5.4494 86.2069 2500 0.7987
5.3702 103.4483 3000 0.7979
5.3229 120.6897 3500 0.7980
5.2742 137.9310 4000 0.7983
5.249 155.1724 4500 0.7990
5.2499 172.4138 5000 0.7993

Framework versions

  • PEFT 0.10.0
  • Transformers 4.41.0.dev0
  • Pytorch 2.2.2+cu121
  • Datasets 2.19.0
  • Tokenizers 0.19.1
Downloads last month
0
Unable to determine this model’s pipeline type. Check the docs .

Adapter for