Abstract
The Kölner Korpus des Kiezdeutschen/Cologne Corpus of Kiezdeutsch is a dataset documenting the urban youth language variety known as Kiezdeutsch as spoken in Cologne (North Rhine-Westphalia), Germany. It includes audio recordings and GAT 2-transcribed conversations among adolescent male speakers recorded in 2023. The data were collected in a vocational school and comprise approximately three hours of conversation across three speaker groups: monolingual, multilingual, and mixed. The corpus is pseudonymized and published under a CC BY 4.0 license. It is intended as a broadly reusable linguistic resource and provides empirical data for research in sociolinguistics, morphosyntax, grammatical variation, lexical innovation, discourse-pragmatics and interactional linguistics. Its structure and basic annotation also make it suitable for applications in language contact research, corpus-based analysis and language pedagogy.