davidilag/wav2vec2-xls-r-300m-cpt-1000h_faroese-converted-best-faroese-100h-30-epochs_run2_2025-08-21

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

wav2vec2-xls-r-300m-pretrained-1000hfaroese-converted-faroese-100h-30-epochs_2025-08-21

This model was fine-tuned using the Ravnursson Faroese ASR dataset. It achieves the following results on the test set:

• —Loss: 0.0988
• —Wer: 7.57
• —Cer: 2.19

Model description

The model has been fine-tuned with 100h of Faroese on a continously pre trained model with 1000h Fareose data

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: 0.0001
• —trainbatchsize: 16
• —evalbatchsize: 8
• —seed: 42
• —gradientaccumulationsteps: 2
• —totaltrainbatch_size: 32
• —optimizer: Use OptimizerNames.ADAMWTORCHFUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• —lrschedulertype: cosine
• —lrschedulerwarmup_steps: 5000
• —num_epochs: 30
• —mixedprecisiontraining: Native AMP

Training results

Framework versions

• —Transformers 4.55.2
• —Pytorch 2.8.0+cu126
• —Datasets 4.0.0
• —Tokenizers 0.21.4