LAGOS — In a move that could reshape how technology speaks to Africa, Google has teamed up with top African universities and grassroots organizations to launch WAXAL, a powerful open speech dataset designed to help artificial intelligence finally understand — and speak — African languages.

For years, voice assistants and speech-powered tools have exploded across the world, yet most of Africa’s more than 2,000 languages have remained digitally invisible. The reason is simple: AI systems need massive amounts of high-quality speech data to learn, and for many African languages, that data just didn’t exist. The result has been a silent exclusion affecting hundreds of millions of people who cannot fully use voice-driven technology in their mother tongues.

WAXAL is aiming to change that story. Built over three years with Google’s funding and technical support, the project has produced 1,250 hours of transcribed natural speech and over 20 hours of studio-quality recordings across 21 Sub-Saharan African languages, including Yoruba, Hausa, Igbo, Swahili, Luganda, and Acholi. The dataset lays the groundwork for AI tools that could potentially serve more than 100 million speakers.

“This is about empowerment,” said Aisha Walcott-Bryant, Head of Google Research Africa, describing WAXAL as a turning point for homegrown innovation.

According to her, the project equips African students, researchers, and developers with the resources to build technology on their own terms, in their own languages — from education apps to voice-powered services that can open doors to real economic opportunities.

Unlike past tech initiatives often criticized for extracting African data without local control, WAXAL was built through direct partnerships with institutions that led the work themselves. Universities and community organizations such as Makerere University in Uganda, the University of Ghana, and Digital Umuganda in Rwanda managed the data collection, while Google provided technical backing.

Crucially, the institutions retain full ownership of the data, setting what many see as a new benchmark for ethical and equitable AI development on the continent.
Researchers say the impact is already visible. At Makerere University, the project has strengthened local AI research and sparked student-led innovations. “For AI to truly work in Africa, it has to understand our languages and realities,” said Joyce Nakatumba-Nabende, a senior lecturer at the university. “WAXAL has already helped us move closer to that goal.”

At the University of Ghana, the initiative has grown beyond a research effort into a broader movement, energizing students and faculty around the idea that Africa can shape the future of AI, rather than just consume it.

With WAXAL now open, the hope is that Africa’s linguistic diversity — long seen as a barrier in the digital age — could become a strength, powering a new wave of inclusive technology built by Africans, for Africans.