wav2vec2-large-xls-r-300m-georgian-large

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the common_voice_10_0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.4291
• Wer: 0.6392

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: 15
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
3.0867	4.21	400	3.1211	1.0
2.8871	8.42	800	2.2250	1.0
0.3667	12.63	1200	0.4291	0.6392

Framework versions

• Transformers 4.11.3
• Pytorch 1.10.0+cu113
• Datasets 2.4.0
• Tokenizers 0.10.3