Leyu: Crowdsourcing Datasets for Ethiopian Languages

High-quality, diverse datasets are the cornerstone of successful AI development. They enable AI systems to understand language nuances, analyze data effectively, and provide solutions tailored contextually. However, the lack of such datasets for Ethiopian languages and industries has hindered AI's potential impact by creating information gaps and language barriers among users. 

Leyu (ለዩ), which translates to identify or label in Amharic, addresses this challenge. Our platform is designed to crowdsource datasets for low-resource languages, focusing on Ethiopia's unique linguistic landscape.

Leyu is building comprehensive datasets for Amharic, Afaan Oromo, Tigrigna, and other low-resource local languages. Our initial focus is on speech data, with plans to expand to video and image datasets. By leveraging the widespread use of smartphones, Leyu democratizes the data creation process and provides micro-work opportunities for Ethiopians to contribute to AI development, ensuring fair compensation. Leyu's crowdsourcing model, which involves local data creators, annotators, and reviewers, ensures that datasets are accurate and diverse, minimizing errors and biases. We use a hybrid human and automated labeling approach to create large, accurate datasets efficiently and cost-effectively. Our datasets will power AI applications in key sectors like agriculture, healthcare, and education. 

We envision a future where AI technology is accessible and beneficial to all Ethiopians. By building high-quality datasets, we aim to enhance AI's impact in agriculture, health, education, and other critical sectors. We also strive to create job opportunities drive economic development through micro-work, and foster relevant AI solutions that address local challenges. 

Leyu is currently in its testing phase, and we're excited about the potential to revolutionize AI in Ethiopia. Learn more at leyu.ai.

Stay tuned for our official launch.