Underwater Vessel Noise Challenge by the French Embassy in South Africa supports machine-learning participants with prizes for models and research that help marine scientists identify vessel noise off South Africa.
Funder: Zindi
Due Dates: December 13, 2026 (Submission close) | December 13, 2026 (Submission selection close) | January 29, 2027 (Results reveal)
Funding Amounts: $5,000 USD across five cash prizes, plus a first-place summit invitation worth $2,500 USD
Summary: A machine-learning prize competition supporting detection and interpretation of vessel-related underwater noise along South Africa’s coast.
Key Information: The top three leaderboard winners must assign worldwide copyright in their winning solution code to Zindi.
Hosted on Zindi, this challenge rewards machine-learning solutions and research write-ups that help marine scientists distinguish vessel-related underwater sound from wind, waves and marine life. The French Embassy in South Africa, Lesotho and Malawi is the main sponsor through Le Shift Lab, with SAEON also sponsoring. Participants use real hydrophone measurements supplied by SAEON, SANBI and the SA Underwater Radiated Noise Consortium to develop or adapt models for South African coastal conditions.
The competition combines automated prediction of known vessel-sound additions with separate research prizes for evidence on real South African measurements. Research write-ups should produce useful pressure maps showing variation across monitoring sites and time, supported by reproducible acoustic evidence, controls and uncertainty. Synthetic test performance alone does not establish reliable monitoring of natural vessel activity.