wav2vec2-large-xls-r-300m-urdu-colab

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the audiofolder dataset. It achieves the following results on the evaluation set:

• Loss: inf
• Wer: 0.4194

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: 0.0003
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
6.1372	3.09	400	inf	0.8798
0.8691	6.18	800	inf	0.5099
0.4049	9.27	1200	inf	0.4904
0.2474	12.36	1600	inf	0.4523
0.1633	15.44	2000	inf	0.4285
0.1164	18.53	2400	inf	0.4194

Framework versions

• Transformers 4.32.0.dev0
• Pytorch 2.0.1+cu118
• Datasets 1.18.3
• Tokenizers 0.13.3