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: 1.2695
• Accuracy: 0.6429

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: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	1	1.6348	0.1429
No log	2.0	2	1.6348	0.1429
No log	3.0	3	1.6132	0.1429
No log	4.0	4	1.5740	0.2857
No log	5.0	5	1.5081	0.3571
No log	6.0	6	1.4480	0.5
No log	7.0	7	1.3837	0.5
No log	8.0	8	1.3302	0.6429
No log	9.0	9	1.2896	0.6429
No log	10.0	10	1.2695	0.6429

Framework versions

• Transformers 4.56.1
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.0