Wav2Vec2_Finetuned

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

• Loss: 0.9886
• Accuracy: 0.7247

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: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.1066	0.9863	54	2.0550	0.3211
1.7945	1.9817	108	1.8463	0.3858
1.5042	2.9772	162	1.6106	0.4911
1.3307	3.9909	217	1.3656	0.6199
1.1295	4.9863	271	1.2506	0.6417
1.0127	5.9817	325	1.2754	0.6211
0.949	6.9772	379	1.0925	0.7041
0.8618	7.9909	434	1.0693	0.7052
0.7838	8.9863	488	1.0308	0.7138
0.7813	9.9452	540	0.9886	0.7247

Framework versions

• Transformers 4.46.2
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3