The University of the Witwatersrand hosted an artificial intelligence and music showcase at its Johannesburg campus April 16, bringing together five teams of musicians and technologists from across the continent to demonstrate AI tools developed specifically for African musical traditions.
The event, organized by the Wits Innovation Centre and the Machine Intelligence and Neural Discovery Institute, drew participants from South Africa, Ghana, Cameroon, Kenya and Ethiopia. Each team paired an artist with an AI specialist and spent several months developing culturally grounded tools, which were presented through live demonstrations and discussions.
Joshua Kroon and Emmanuel Apetsi introduced the Bebii Engine, an AI-driven music generation system aimed at preserving indigenous knowledge through sound. Their demonstration included a live performance of a traditional African composition using indigenous instruments, with the system processing and adapting the material in real time.
A second project, Timah.AI, developed by Tora Nyamosi and Lawrence Moruye, focused on preservation through digitization. The web-based platform allows users to upload recordings of traditional music, which are then transcribed and stored in a searchable digital archive intended to support access to African musical heritage.
Linda Nyabundi and Gebregziabihier Niguise presented Heritage in Code, a response to the limited presence of African music datasets on mainstream AI platforms. The project centers on building structured datasets enriched with metadata and cultural context, aimed at improving how AI systems recognize and generate African musical forms.
One of the most widely received presentations came from South African artist Umlilo and Ghanaian engineer Gideon Gyimah, who developed Zazi, an AI-powered co-creation tool designed for musicians. The system includes stem isolation, voice cloning, track merging and AI-assisted mastering, with a focus on African languages and tonal systems. During a live demonstration, the platform generated a maskandi-inspired track with isiZulu vocal elements, responding to multilingual prompts in real time.
The final project, Bina.AI, developed by Ehinome Ogbeide and Ashuza Muhigiri, focused on early childhood learning. The platform generates personalized songs and stories for children embedded with African cultural contexts and educational content.
The program also included a keynote address on the state of generative AI in music and a roundtable discussion on creative sovereignty in African music and AI, featuring industry professionals and researchers.
Across all five presentations, a shared emphasis emerged on cultural preservation, accessibility and the practical challenges of integrating AI into African music ecosystems.
