drelhaj/Arabic-Dialects

Arabic Dialects Dataset (Bivalency & Code-Switching) The Arabic Dialects Dataset is a specialised corpus designed for automatic dialect identification, with a focus on the linguistic phenomena of bivalency and written code-switching between major Arabic dialects and Modern Standard Arabic (MSA).It covers five varieties: EGY – Egyptian Arabic GLF – Gulf Arabic LAV – Levantine Arabic NOR – North African / Tunisian Arabic MSA – Modern Standard Arabic The dataset was… See the full description on the dataset page: https://huggingface.co/datasets/drelhaj/Arabic-Dialects.