This dataset contains speech segments in Tamazight (specifically focusing on the Tachelhit dialect) paired with their corresponding Arabic transcriptions. It is designed to support the development of automatic speech recognition (ASR) systems for the Tamazight language, particularly for translation into Modern Standard Arabic.
This is an actively growing dataset, with regular updates and new data points being… See the full description on the dataset page: https://huggingface.co/datasets/SoufianeDahimi/Tamazight-ASR-Dataset-v2.
