Model_G_Wav2Vec2

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

• Loss: 0.4452
• Wer: 0.3653
• Cer: 0.0942

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	Cer
3.759	5.97	400	0.7276	0.7372	0.2083
0.3491	11.94	800	0.5168	0.4853	0.1312
0.1556	17.91	1200	0.4654	0.4210	0.1117
0.0981	23.88	1600	0.4576	0.3896	0.1020
0.065	29.85	2000	0.4452	0.3653	0.0942

Framework versions

• Transformers 4.11.3
• Pytorch 1.10.0+cu113
• Datasets 1.18.3
• Tokenizers 0.10.3