Building the data infrastructure layer for African language AI
Afriklang builds the data infrastructure for African language AI commercially licensed, speaker-verified speech datasets for the AI and CPaaS teams building products in African languages, starting with Twi, Wolof, and Fon and scaling on demand.
What we stand for
Commercial-grade by default
Every dataset ships with a commercial license and an SLA no legal grey zones when you deploy at scale.
Fair-Trade sourcing
Native speakers are compensated fairly through our points-based micro-work system. Good data starts with fair work.
Linguistic rigor
Every annotation is cross-checked to a >80% inter-annotator agreement standard, overseen by native linguists.
How we build the data
Every dataset runs through the same controlled pipeline quality is engineered in at capture, not patched at the end.
Image-prompted elicitation
Native speakers describe images instead of reading scripts, capturing natural, code-switched speech.
AI quality-filtering at capture
Noisy or low-quality audio is discarded automatically before it ever reaches a human reviewer.
Native annotation & IAA cross-check
Every hour is transcribed and labeled by native speakers, cross-checked to a >80% inter-annotator agreement standard.
Benchmarked & delivered
We evaluate against Whisper and MMS baselines and deliver via API or S3 under a commercial license and SLA.
The people behind Afriklang
A team spanning operations, business, AI and linguistics building within the MEST Africa ecosystem.
Agnilonda Pakou
Team Lead
Operations
Khalifa Niamadio Mamadou
Business Lead
Business
Mahougnon Fredy Houndayi
AI Lead
AI & Engineering
Babacar Ndao
Linguistic Lead
Linguistics
A MEST Africa company
Afriklang is built and backed within the MEST Africa ecosystem one of the continent's leading technology entrepreneur programs and investors.
Let's build African-language AI together.
Or email us at contact@afriklang.com