The WASABI Dataset: Cultural, Lyrics and Audio Analysis Metadata About 2 Million Popular Commercially Released Songs

Since 2017, the goal of two-million song WASABI database has been to build a knowledge graph linking collected metadata (artists, discography, producers, dates, etc.) with generated by analysis both songs’ lyrics (topics, places, emotions, structure, and audio signal (chords, sound, etc.). It relies on natural language processing machine learning methods for extraction, semantic Web frameworks forrepresentation integration. describes more than 2 millions commercial songs, 200K albums 77K artists. can be exploited music search engines, professionals (e.g. journalists, radio presenters, teachers) or scientists willing analyze popular published since 1950. is available under an open license, in multiple formats online source services including interactive navigator, REST API SPARQL endpoint.