ACTS - AI4D DHOLUO PROJECT
AI for climate resilience and farmer advisory systems
Overview
The main goal of this project is to develop a model for transcribing dholuo speech to text and text to speech conversations. The main focus is towards the agricultural context of the Luo community.
Project logic
Aim
Our Mission is to harness Artificial Intelligence to preserve, promote and advance the Dholuo language, Culture and Heritage through communication, education, research and digital inclusion. We aim to develop a model for transcribing dholuo speech to text and text to speech conversations, with a focus on the agricultural context of the Luo community.
Context
Smallholder farmers lack reliable, local-language access to agronomic guidance, market prices and climate-adaptive systems that they can communicate with in their own language; access barriers disproportionately affect women and remote communities.
Tasks
Collect multilingual speech and text data, build translation and language-ID components, develop a field-facing advisory pipeline, and integrate with local extension services.
Success
A deployed advisory prototype that delivers voice and text guidance in local languages, measurable increases in farmer access to recommendations, and documentation for scale-up.
Focus areas
- Agriculture
- Speech Technology
- Machine Translation
- Data Collection