audio_cls_nkn

This model is a fine-tuned version of Kkonjeong/wav2vec2-base-korean on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6969
• Accuracy: 0.8992

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	15	2.3611	0.3529
No log	2.0	30	2.0563	0.5126
No log	3.0	45	1.6787	0.6555
No log	4.0	60	1.3515	0.7563
No log	5.0	75	1.2295	0.7311
No log	6.0	90	0.9776	0.8487
No log	7.0	105	0.8444	0.8655
No log	8.0	120	0.7801	0.8824
No log	9.0	135	0.7146	0.8824
No log	10.0	150	0.6969	0.8992

Framework versions

• Transformers 4.57.1
• Pytorch 2.8.0+cu126
• Datasets 3.6.0
• Tokenizers 0.22.1