wav2vec2-telugu

This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the openslr dataset. It achieves the following results on the evaluation set:

• Loss: 0.2982
• Wer: 0.2885

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: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
6.6905	3.84	400	0.7109	0.7800
0.4532	7.69	800	0.2972	0.3977
0.1957	11.54	1200	0.2907	0.3522
0.1284	15.38	1600	0.3117	0.3317
0.0979	19.23	2000	0.3000	0.3353
0.0749	23.08	2400	0.2823	0.3045
0.0584	26.92	2800	0.2982	0.2885

Framework versions

• Transformers 4.24.0
• Pytorch 1.13.0+cu117
• Datasets 2.6.1
• Tokenizers 0.13.2