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

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: 0.3755
• Wer: 0.3054

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
3.9475	3.67	400	0.7146	0.7507
0.4396	7.34	800	0.4402	0.4574
0.1955	11.01	1200	0.4027	0.4056
0.128	14.68	1600	0.4127	0.3826
0.0977	18.35	2000	0.4156	0.3576
0.0788	22.02	2400	0.3992	0.3394
0.0637	25.69	2800	0.3801	0.3141
0.0486	29.36	3200	0.3755	0.3054

Framework versions

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