wav2vec2-large-xlsr-faroese-5k-steps

This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1857
• Wer: 0.3436

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
• training_steps: 5000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.7004	0.4640	1000	0.5430	0.6687
0.456	0.9281	2000	0.3044	0.4647
0.322	1.3921	3000	0.2425	0.3967
0.2624	1.8561	4000	0.1946	0.3530
0.1986	2.3202	5000	0.1857	0.3436

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.0
• Tokenizers 0.19.1