This dataset contains augmented audio samples with pitch shifting and white noise enhancement for improved model training and robustness.
This is an augmented version of the original audio dataset, processed with audio augmentation techniques to increase dataset diversity and improve model generalization.
Pitch Shifting: Random pitch shift between -2 to +2 semitones
White Noise… See the full description on the dataset page:
https://huggingface.co/datasets/milanakdj/augmented-merged-and-shuffel-tibetan-dataset.
