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

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the common_voice_6.1 dataset. It achieves the following results on the evaluation set:

• Loss: 0.4380
• Wer: 0.3508

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
3.8764	3.67	400	0.7239	0.7221
0.4526	7.34	800	0.5009	0.5345
0.2169	11.01	1200	0.4728	0.4693
0.1438	14.68	1600	0.4648	0.4669
0.1095	18.35	2000	0.4642	0.4094
0.0893	22.02	2400	0.4749	0.3879
0.0701	25.69	2800	0.4410	0.3665
0.056	29.36	3200	0.4380	0.3508

Framework versions

• Transformers 4.11.3
• Pytorch 1.10.0+cu111
• Datasets 1.18.3
• Tokenizers 0.10.3