Crowdsourcing datasets
For low-resource languages across Africa.
Fueling innovation tailored to local needs!
Our Approach
Crowdsourcing
Collecting data from local annotators, ensuring cultural accuracy and diverse datasets.
Social impact and fair Wages
Creating micro-work opportunities and ensuring fair wages for youth and Women.
Hybrid Labeling
Combining human and automated labeling for accurate datasets.
Segmented Data
Creating distinct training, validation, and test sets for LLMs, enhancing adaptability across sectors.
Data Privacy & Ethics
Ensuring ethical dataset through legal compliance and data ownership respect.
Marketplace
Providing a platform for companies to purchase datasets, facilitating access to high-quality data.
Why Leyu?
Inclusive datasets
We move beyond generic datasets, offering accurate data in Amharic, Afaan Oromoo, and Tigrigna.
Empowers local annotators
Through a crowdsourcing platform, Leyu empowers local annotators, especially women and youth to contribute their voices particularly in AI.
Ethical data use
We champion ethical data use and fair compensation to fuel local innovation designed for local challenges.
Language
Preserving and amplifying Ethiopian voices by collecting data in their native languages, ensuring their voices are heard and reflected in dataset development.
What We Do
Our Values
Data Quality and Accuracy
Diverse and high-quality data for reliable AI performance.
Accessibility
Accessible and cost-effective solutions for various sectors.
Impact
Solutions addressing societal challenges in agriculture, health, education and beyond.
Collaboration and Partnership
Sector-specific language datasets for key industries like agriculture, health, and education and beyond.
