Downloads · 30 days
13
37% of all-time downloads
korean_kws
alwaysnotavailable/korean_kws
korean_kws is a audio classification model from alwaysnotavailable. Use it for the audio classification task on the model card, and read the license before you ship it in a product. It is set up for transformers.
should probably proofread and complete it, then remove this comment. --
All-time downloads
35
Public
Parameters
94.6M
378 MB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors378 MB · 100%
At a glance
- Task
- Audio Classification
- Library
- transformers
- Model type
- wav2vec2
- Access
- Public
- Created
- Jan 5, 2026
- Updated
- Jan 5, 2026
- SHA
- 647d1ab9
Base models
Model card
From the Hugging Face model README
<!-- 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_kws
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: 0.3329
- 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 | 1.9572 | 0.0769 |
| No log | 2.0 | 4 | 1.9177 | 0.2308 |
| No log | 3.0 | 6 | 1.8590 | 0.3077 |
| No log | 4.0 | 8 | 1.7834 | 0.3846 |
| No log | 5.0 | 10 | 1.7145 | 0.3846 |
| No log | 6.0 | 12 | 1.5883 | 0.6154 |
| No log | 7.0 | 14 | 1.4392 | 0.9231 |
| No log | 8.0 | 16 | 1.3025 | 1.0 |
| No log | 9.0 | 18 | 1.1472 | 1.0 |
| No log | 10.0 | 20 | 1.0278 | 1.0 |
| No log | 11.0 | 22 | 0.9255 | 1.0 |
| No log | 12.0 | 24 | 0.8274 | 1.0 |
| No log | 13.0 | 26 | 0.7610 | 1.0 |
| No log | 14.0 | 28 | 0.6741 | 1.0 |
| No log | 15.0 | 30 | 0.6269 | 1.0 |
| No log | 16.0 | 32 | 0.6036 | 1.0 |
| No log | 17.0 | 34 | 0.5516 | 1.0 |
| No log | 18.0 | 36 | 0.5066 | 1.0 |
| No log | 19.0 | 38 | 0.4697 | 1.0 |
| No log | 20.0 | 40 | 0.4436 | 1.0 |
| No log | 21.0 | 42 | 0.4216 | 1.0 |
| No log | 22.0 | 44 | 0.4011 | 1.0 |
| No log | 23.0 | 46 | 0.3842 | 1.0 |
| No log | 24.0 | 48 | 0.3702 | 1.0 |
| No log | 25.0 | 50 | 0.3588 | 1.0 |
| No log | 26.0 | 52 | 0.3501 | 1.0 |
| No log | 27.0 | 54 | 0.3434 | 1.0 |
| No log | 28.0 | 56 | 0.3383 | 1.0 |
| No log | 29.0 | 58 | 0.3348 | 1.0 |
| No log | 30.0 | 60 | 0.3329 | 1.0 |
Framework versions
- Transformers 4.57.3
- Pytorch 2.9.0+cu126
- Datasets 4.0.0
- Tokenizers 0.22.1
- Task
- Audio Classification
- Library
- transformers
- Type
- wav2vec2
- Created
- Jan 5, 2026
- Updated
- Jan 5, 2026