EYEDOL/Yoruba-ASR3

YorubaASR3

This model is a fine-tuned version of EYEDOL/Yoruba-ASR2 on the Common Voice 17.0 dataset. It achieves the following results on the evaluation set:

• Loss: 1.0831
• Wer: 42.3035

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: 1e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0001	15.3846	4000	1.0831	42.3035

Framework versions

• Transformers 4.41.2
• Pytorch 2.1.2
• Datasets 3.2.0
• Tokenizers 0.19.1