ezra-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: 0.3891
• Wer: 0.3459

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

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
3.8755	3.67	400	0.6685	0.7130
0.4087	7.34	800	0.4413	0.4598
0.1917	11.01	1200	0.4143	0.4399
0.1318	14.68	1600	0.4082	0.3908
0.1035	18.35	2000	0.4184	0.3926
0.0801	22.02	2400	0.3847	0.3751
0.0604	25.69	2800	0.4148	0.3591
0.0501	29.36	3200	0.3891	0.3459

Framework versions

• Transformers 4.34.0
• Pytorch 2.1.0+cu121
• Datasets 2.14.5
• Tokenizers 0.14.1