Skip to content
ModelListedExternal
Download

Korean Kws2

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: 0.3841 Accuracy: 1.0

Sourcehuggingface.co/seohyeon122333/korean_kws2/tree/d109b2028268166935577ec283e5d83242e32cae
KindOtherBase modelKkonjeong/wav2vec2-base-koreanVersionRevision d109b20Cgrade

Model source

Kind
Other
File size
361 MB
Base model
Kkonjeong/wav2vec2-base-korean
Version
vRevision d109b20
Updated
Oct 17, 2026
Downloads
5
Source
Hugging Face

Source description

korean_kws2

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: 0.3841
  • 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 OptimizerNames.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 log1.021.96080.0
No log2.041.96890.0
No log3.061.96890.0
No log4.081.92830.3077
No log5.0101.90700.5385
No log6.0121.83160.4615
No log7.0141.70730.4615
No log8.0161.57590.5385
No log9.0181.46180.7692
No log10.0201.32501.0
No log11.0221.18281.0
No log12.0241.06830.9231
No log13.0260.96771.0
No log14.0280.86941.0
No log15.0300.81571.0
No log16.0320.74031.0
No log17.0340.67801.0
No log18.0360.62771.0
No log19.0380.59661.0
No log20.0400.55721.0
No log21.0420.52151.0
No log22.0440.48261.0
No log23.0460.45681.0
No log24.0480.43701.0
No log25.0500.42001.0
No log26.0520.40741.0
No log27.0540.39781.0
No log28.0560.39051.0
No log29.0580.38611.0
No log30.0600.38411.0

Framework versions

  • Transformers 5.0.0
  • Pytorch 2.10.0+cu128
  • Datasets 4.0.0
  • Tokenizers 0.22.2

Sources

1 source
Hugging FacePrimaryVersion d109b2028268166935577ec283e5d83242e32cae
huggingface.co/seohyeon122333/korean_kws2/tree/d109b2028268166935577ec283e5d83242e32cae

Verified Aug 29

Model artifacts

1 artifact

model.safetensors

safetensors · 361 MB · SHA-256 24ef6ab54b3a…faf0 · Hugging Face

Download file

Source excerpts

2 excerpts
ReadmeAug 29huggingface.co/seohyeon122333/korean_kws2

--- library_name: transformers base_model: Kkonjeong/wav2vec2-base-korean tags: - generated_from_trainer metrics: - accuracy model-index: - name: korean_kws2 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. --> # korean_kws2 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: 0.3841 - 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 OptimizerNames.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 | 1.9608 | 0.0 | | No log | 2.0 | 4 | 1.9689 | 0.0 | | No log | 3.0 | 6 | 1.9689 | 0.0 | | No log | 4.0 | 8 | 1.9283 | 0.3077 | | No log | 5.0 | 10 | 1.9070 | 0.5385 | | No log | 6.0 | 12 | 1.8316 | 0.4615 | | No log | 7.0 | 14 | 1.7073 | 0.4615 | | No log | 8.0 | 16 | 1.5759 | 0.5385 | | No log | 9.0 | 18 | 1.4618 | 0.7692 | | No log | 10.0 | 20 | 1.3250 | 1.0 | | No log | 11.0 | 22 | 1.1828 | 1.0 | | No log | 12.0 | 24 | 1.0683 | 0.9231 | | No log | 13.0 | 26 | 0.9677 | 1.0 | | No log | 14.0 | 28 | 0.8694 | 1.0 | | No log | 15.0 | 30 | 0.8157 | 1.0 | | No log | 16.0 | 32 | 0.7403 | 1.0 | | No log | 17.0 | 34 | 0.6780 | 1.0 | | No log | 18.0 | 36 | 0.6277 | 1.0 | | No log | 19.0 | 38 | 0.5966 | 1.0 | | No log | 20.0 | 40 | 0.5572 | 1.0 | | No log | 21.0 | 42 | 0.5215 | 1.0 | | No log | 22.0 | 44 | 0.4826 | 1.0 | | No log | 23.0 | 46 | 0.4568 | 1.0 | | No log | 24.0 | 48 | 0.4370 | 1.0 | | No log | 25.0 | 50 | 0.4200 | 1.0 | | No log | 26.0 | 52 | 0.4074 | 1.0 |...

NoteAug 29huggingface.co/seohyeon122333/korean_kws2

Source context: 5 downloads · 0 likes · Pipeline audio-classification · Library transformers · Repo seohyeon122333/korean_kws2