Skip to content
ModelListedExternal
Download

Source

This model is a fine-tuned version of Kkonjeong/wav2vec2-base-korean on the None dataset. It achieves the following results on the evaluation set: Loss: 0.3088 Accuracy: 1.0

Sourcehuggingface.co/hshshsuuu/source/tree/f773ea4832ac36953f714eb0b55b6626fbf74fe7
KindOtherBase modelKkonjeong/wav2vec2-base-koreanVersionRevision f773ea4

Model source

Kind
Other
File size
361 MB
Base model
Kkonjeong/wav2vec2-base-korean
Version
vRevision f773ea4
Updated
Sep 13, 2026
Downloads
2
Source
Hugging Face

Source description

source

This model is a fine-tuned version of Kkonjeong/wav2vec2-base-korean on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3088
  • Accuracy: 1.0

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: 0.0001
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 32
  • optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • num_epochs: 30
  • mixed_precision_training: Native AMP

Training results

Training LossEpochStepValidation LossAccuracy
No log1.021.84260.7895
No log2.041.78520.4211
No log3.061.70760.4211
No log4.081.56830.4737
No log5.0101.40960.5263
No log6.0121.26350.6842
No log7.0141.15450.8947
No log8.0161.06351.0
No log9.0181.03641.0
No log10.0200.93861.0
No log11.0220.82131.0
No log12.0240.73241.0
No log13.0260.67621.0
No log14.0280.61861.0
No log15.0300.56641.0
No log16.0320.52221.0
No log17.0340.47901.0
No log18.0360.44661.0
No log19.0380.41821.0
No log20.0400.39771.0
No log21.0420.38531.0
No log22.0440.36291.0
No log23.0460.41651.0
No log24.0480.42380.9474
No log25.0500.34181.0
No log26.0520.33161.0
No log27.0540.32131.0
No log28.0560.31571.0
No log29.0580.31111.0
No log30.0600.30881.0

Framework versions

  • Transformers 4.57.6
  • Pytorch 2.10.0+cu128
  • Datasets 4.5.0
  • Tokenizers 0.22.2

Sources

1 source
Hugging FacePrimaryVersion f773ea4832ac36953f714eb0b55b6626fbf74fe7
huggingface.co/hshshsuuu/source/tree/f773ea4832ac36953f714eb0b55b6626fbf74fe7

Verified Sep 13

Model artifacts

1 artifact

model.safetensors

safetensors · 361 MB · SHA-256 8772bb65a07a…f917 · Hugging Face

Download file

Source excerpts

2 excerpts
ReadmeSep 13huggingface.co/hshshsuuu/source

--- library_name: transformers base_model: Kkonjeong/wav2vec2-base-korean tags: - generated_from_trainer metrics: - accuracy model-index: - name: source results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # source This model is a fine-tuned version of [Kkonjeong/wav2vec2-base-korean](https://huggingface.co/Kkonjeong/wav2vec2-base-korean) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.3088 - Accuracy: 1.0 ## 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: 0.0001 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 32 - optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - num_epochs: 30 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | |:-------------:|:-----:|:----:|:---------------:|:--------:| | No log | 1.0 | 2 | 1.8426 | 0.7895 | | No log | 2.0 | 4 | 1.7852 | 0.4211 | | No log | 3.0 | 6 | 1.7076 | 0.4211 | | No log | 4.0 | 8 | 1.5683 | 0.4737 | | No log | 5.0 | 10 | 1.4096 | 0.5263 | | No log | 6.0 | 12 | 1.2635 | 0.6842 | | No log | 7.0 | 14 | 1.1545 | 0.8947 | | No log | 8.0 | 16 | 1.0635 | 1.0 | | No log | 9.0 | 18 | 1.0364 | 1.0 | | No log | 10.0 | 20 | 0.9386 | 1.0 | | No log | 11.0 | 22 | 0.8213 | 1.0 | | No log | 12.0 | 24 | 0.7324 | 1.0 | | No log | 13.0 | 26 | 0.6762 | 1.0 | | No log | 14.0 | 28 | 0.6186 | 1.0 | | No log | 15.0 | 30 | 0.5664 | 1.0 | | No log | 16.0 | 32 | 0.5222 | 1.0 | | No log | 17.0 | 34 | 0.4790 | 1.0 | | No log | 18.0 | 36 | 0.4466 | 1.0 | | No log | 19.0 | 38 | 0.4182 | 1.0 | | No log | 20.0 | 40 | 0.3977 | 1.0 | | No log | 21.0 | 42 | 0.3853 | 1.0 | | No log | 22.0 | 44 | 0.3629 | 1.0 | | No log | 23.0 | 46 | 0.4165 | 1.0 | | No log | 24.0 | 48 | 0.4238 | 0.9474 | | No log | 25.0 | 50 | 0.3418 | 1.0 | | No log | 26.0 | 52 | 0.3316 | 1.0 | | No log | 27.0 | 54...

NoteSep 13huggingface.co/hshshsuuu/source

Source context: 2 downloads · 0 likes · Pipeline audio-classification · Library transformers · Repo hshshsuuu/source