Togo launches National Local Language Challenge to train AI models for digital public services

Date: 2026-10-08
news-banner

By:   Robert Kwaku Annor

Togo has launched the National Local Language Challenge, opening a contribution platform to collect texts, recordings and translations that will help artificial intelligence models understand the country’s local languages and improve access to digital public services.

The initiative was launched in Lomé by Cina Lawson, Minister of Public Service Efficiency and Digital Transformation, as the first stage of the #VoixDuTogo program.

The launch was attended by several members of the Government, including Colonel Hodabalo Awate, Minister of Territorial Administration and Customary Affairs; Isaac Tchiakpe, Minister of Culture and Tourism; and Prof Gado Tchangbedji, Minister Delegate for Higher Education and Scientific Research.

The initiative seeks to address the limited availability of digital language resources in Togo by building datasets that can be used to train and improve AI models.

“In order for a citizen to be able to communicate with a digital public service in his or her language in the future, the Intelligence Artificielle must learn to understand it,” the ministry said.

To achieve this, the program is seeking texts, recordings and translations in local languages that are currently not sufficiently represented in digital resources.

The first phase will run as a one-month pilot covering Ewe, Kabiyè and Moba. The pilot will be used to test the approach before it is extended to other languages spoken in Togo.

Participants can create an account using their phone number and indicate the languages they speak, read or write. They will then receive tasks adapted to their skills, including reading sentences aloud, transcribing recordings, translating texts and verifying contributions from other participants.

The collected data will be validated by linguists and local communities before being used to support AI development.

The Togo AI Lab, together with students, researchers and developers, will contribute to exploiting the resources to train and improve AI models, including through competitions organized on Zindi.

The government says the initiative is aimed at advancing digital inclusion through tools that can facilitate access to public services, health and justice in local languages.

Field agents will also collect contributions in localities where access to telephone and internet services remains limited, allowing communities with limited connectivity to participate in the development of the country's digital language resources.

The initiative is intended to build these resources collectively, with contributions from citizens providing examples that can help future AI models better understand Togo’s linguistic diversity.

Leave Your Comments