Meet Mutembesa and Gloria – Using Machine Learning to Improve Their Communities
25 April 2024
In Uganda and Nigeria, two grantees tackle livestock disease diagnostics and audio dataset creation for native Nigerian languages
In Uganda, data scientist and farmer Mutembesa Daniel is working to more effectively diagnose, track and manage livestock disease and spread. Through his team’s efforts, data collected by veterinarians in remote villages will be able to be routed to a national dashboard for livestock agencies to utilize. “[This] helps precision agriculture target a small, limited resource to exactly where it’s needed,” says Daniel.
In Nigeria, Professor of English Gloria Monica Tobechukwu Emezue (popularly known as “GMT Emezue”) is working with her team to create 1,500 audio datasets in Igbo, Hausa, and Yoruba – 500 for each language. By providing audio datasets in these three Nigerian languages, the team will enable voice technologies that can be applied in many sectors. “[The datasets] could be used in the education sector, health sector, in agriculture, in fact in every area of life – because language cuts across every area of life and all age groups,” says Emezue.
Mutembesa and Gloria are two Lacuna Fund grantees in our Agriculture and Language domains. Watch both of them below speak to the importance of their work, the datasets they are building, and what funding from Lacuna Fund has enabled them to do.
Note: Both videos have English, Spanish, and French subtitles. Please click the closed captioning button (CC) to view in your preferred language.
Gloria and Mutembesa are two representatives of the unique, trailblazing machine learning datasets being created by our grantees across all our domains, such as datasets for pest and disease diagnosis in crops, or a multi-lingual, Nigerian Twitter sentiment corpus dataset. Agriculture and Natural Language Processing (NLP) were Lacuna Fund’s first two funding domains in 2020 and to-date make up 55% of our funded projects.
Gloria and Mutembesa’s teams are currently building these datasets. Anyone subscribed to this newsletter will be notified when their datasets – or any new Lacuna Fund datasets – are released! If you are sharing this newsletter with a friend, they can subscribe with this link.
We extend our gratitude to the team at Emergence Creative for creating these videos. We are also thankful to Kigali Forward for shooting these interviews, along with many other grantee interviews and photos, at the 2023 Grantee Convening in Rwanda. Finally, we thank Virtual Words for their subtitling translation of these videos.
We extend our thanks, as always, to our funders. Our 2021 Agriculture and Language cohorts are funded by Canada’s International Development Research Centre (IDRC), Google.org, and The Rockefeller Foundation.
“Lacuna Fund is creating lots of opportunities for, not just dataset creation, but [for] new technologies to come into Africa – technologies that will address the African problems in the languages, and in ways, Africans understand.”
– Professor Gloria Monica Tobechukwu Emezue
