audio_cls1

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: 2.0407
• Accuracy: 0.5546

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: 3e-05
• 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	2.6522	0.0420
No log	2.0	30	2.6192	0.0672
No log	3.0	45	2.5423	0.2101
No log	4.0	60	2.4174	0.2437
No log	5.0	75	2.3224	0.3697
No log	6.0	90	2.2396	0.3866
No log	7.0	105	2.1535	0.4538
No log	8.0	120	2.1064	0.4790
No log	9.0	135	2.0601	0.5042
No log	10.0	150	2.0407	0.5546

Framework versions

• Transformers 5.0.0
• Pytorch 2.11.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2