wav2vec2-base-fine-tuning_23_colab

This model is a fine-tuned version of jonatasgrosman/wav2vec2-large-xlsr-53-arabic on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7692
• Wer: 0.5937

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: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
40.2869	1.3333	250	3.3969	1.0
3.314	2.6667	500	2.9060	1.0
1.7031	4.0	750	1.1416	0.7329
0.8774	5.3333	1000	0.8753	0.6403
0.6796	6.6667	1250	0.8064	0.6110
0.5698	8.0	1500	0.7932	0.6147
0.4989	9.3333	1750	0.7692	0.5937

Framework versions

• Transformers 4.45.2
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3