Google, in partnership with African research institutions, has launched WAXAL, a project aimed at expanding African participation in artificial intelligence development.
WAXAL is a large-scale, open speech dataset designed to support research and inclusive voice-enabled AI systems across Africa.
Head of Google Research Africa, Aisha Walcott-Bryant, said the initiative would empower African communities by enabling access to locally relevant AI tools.
She said the project would allow students, researchers and entrepreneurs to build voice-based technologies in their own languages.
According to her, the dataset supports innovation capable of reaching more than 100 million people across the continent.
Walcott-Bryant said African innovators would deploy the data to develop educational tools and voice-enabled services with economic value.
She said WAXAL provides foundational speech data for 21 Sub-Saharan African languages, including Hausa, Yoruba, Luganda and Acholi.
According to her, the project addresses a long-standing digital gap caused by the lack of high-quality speech data for Africa’s indigenous languages.
“While voice technologies are widely used globally, limited data has constrained development across Africa’s over 2,000 languages,” she said.
Walcott-Bryant said WAXAL was developed over three years with Google funding and includes 1,250 hours of transcribed natural speech.
She added that the dataset also features over 20 hours of studio-quality recordings for high-fidelity synthetic voice development.
She said African institutions led the data collection process with technical support from Google.
Partner institutions include Makerere University, the University of Ghana and Rwanda’s Digital Umuganda.
According to her, partner institutions retain full ownership of the data, reinforcing a partnership-led model for AI development in Africa.
