Africa remains largely invisible in the data that powers today’s artificial intelligence systems, despite being home to more than 2,000 languages. Most machine learning models are trained primarily on English, Mandarin, and a handful of other dominant global languages.
For millions of Africans, that gap means the next generation of digital tools—from voice assistants to financial apps—may struggle to understand the languages they speak every day.
Nigeria is now taking a step to change that.
The National Information Technology Development Agency (NITDA) has partnered with NKENNEAi, an African language artificial intelligence platform, to build the infrastructure needed to support AI systems trained specifically on African languages.
The collaboration aims to develop scalable translation and language technologies that can power digital services across government, healthcare, finance, and other sectors in Nigeria’s multilingual society.
With more than 500 languages spoken in Nigeria alone, language remains one of the biggest barriers to digital inclusion. The NITDA–NKENNEAi partnership seeks to close that gap by developing AI systems designed to understand the tonal structures, dialects, and linguistic nuances unique to African languages.
NKENNEAi grew out of NKENNE, one of the fastest-growing African language learning platforms.
Originally created to preserve and teach African languages, the platform has grown into a global community with more than 400,000 users learning languages such as Igbo, Yoruba, Swahili, Hausa, Twi, Somali, and Nigerian Pidgin.
As the platform expanded, it generated something increasingly valuable: a growing dataset of African language text, speech, and linguistic patterns.
That dataset became the foundation for NKENNEAi, a multilingual artificial intelligence platform focused on building the underlying infrastructure required for African language AI.
According to Michael Odokara-Okigbo, CEO of NKENNEAi, the evolution from language learning to AI development emerged naturally as the company’s data and linguistic research expanded.
“NKENNE started as a cultural mission to help preserve and teach African languages,” Odokara-Okigbo said. “As our community grew to hundreds of thousands of learners, we realized the data and linguistic insights we were building could power something much bigger. NKENNEAi is about building the infrastructure that allows African languages to exist and scale inside artificial intelligence systems.”
African languages present challenges that many global AI models are not designed to handle.
Many are tonal, meaning that changes in pitch can alter the meaning of words. They also feature regional dialects and contextual variations that require specialized linguistic understanding.
Because most global models lack sufficient training data for these languages, their translations and voice systems often perform poorly.
NKENNEAi is taking a different approach.
The company is building specialized data pipelines and linguistic annotation systems designed specifically for African languages. These include bilingual sentence datasets for machine translation, annotated speech datasets for speech-to-text systems, and tone-aware linguistic tagging designed to preserve meaning across dialects.
The platform also relies on community validation from native speakers, ensuring the language data reflects real-world usage.
Together, these systems support technologies such as AI translation, speech-to-text transcription, text-to-speech voice synthesis, and multilingual APIs that developers and enterprises can integrate directly into digital platforms.
The platform currently focuses on languages including Yoruba, Igbo, Hausa, Swahili, and Nigerian Pidgin, with plans to expand coverage to additional African languages.
NKENNEAi’s development has also attracted international research support.
Through the U.S. National Science Foundation’s Small Business Innovation Research program, funding was awarded to ESM Global Productions, the company behind NKENNEAi, to advance the development of its multilingual translation platform.
In 2024, the company received a $1 million Phase II NSF award to expand its African language translation API and continue developing speech and language models tailored to tonal languages.
The project supports the creation of multilingual translation models, speech-to-text systems trained on African speech datasets, text-to-speech voice models, and scalable APIs that allow African language capabilities to be integrated into digital platforms.
Together, these efforts are helping build one of the largest structured datasets focused specifically on African languages.
Aligning with Nigeria’s AI ambitions
The partnership also aligns with Nigeria’s broader digital strategy led by the Federal Ministry of Communications, Innovation and Digital Economy, headed by Dr. Bosun Tijani.
Before joining government, Tijani co-founded Co-Creation Hub (CcHUB), one of Africa’s most influential technology innovation centers. Since entering public office, his ministry has been pushing initiatives aimed at positioning Nigeria as a global hub for artificial intelligence and digital innovation.
One of those initiatives, the 3 Million Technical Talent (3MTT) program, is focused on training millions of Nigerians in digital and AI-related skills.
Through the collaboration with NKENNEAi, NITDA is exploring how locally developed AI infrastructure can help serve Nigeria’s multilingual population while strengthening the country’s domestic AI capabilities.
Building the workforce behind language AI
The partnership will also focus on developing the workforce needed to sustain African language AI systems.
Planned programs include training AI data annotators, natural language processing engineers, and public-sector technical teams who will support language dataset development and system deployment.
The goal is to ensure that African language AI is not only built for the continent, but built by the people who understand its languages and cultures best.
Why language AI matters
Africa’s digital economy is growing rapidly, but language remains a major barrier to access.
Millions of Africans are more comfortable communicating in indigenous languages than in English. Yet most digital platforms from banking apps to healthcare systems are still designed primarily for English-speaking users.
Without language accessibility, large segments of the population remain excluded from digital services.
AI-powered language infrastructure could help change that by enabling platforms to communicate with users in the languages they speak every day.
Building Africa’s language future
Global technology companies are beginning to recognize the importance of African languages. Firms like Google have recently expanded AI and search support for languages including Yoruba and Hausa.
But while global companies are gradually adding African languages to their systems, NKENNEAi is focused on building infrastructure designed specifically for the continent’s linguistic complexity.
“African languages are deeply tonal, contextual, and culturally rich,” Odokara-Okigbo said. “Building AI that truly understands them requires infrastructure designed specifically for those languages.”
Through its partnership with NITDA, NKENNEAi plans to roll out its technology in stages, beginning with pilot integrations in government systems, expansion into additional languages, and workforce training programs.
The long-term goal is to build a national infrastructure layer for African language AI, positioning Nigeria as a leader in one of the most overlooked areas of global artificial intelligence.
For NKENNEAi, the mission goes beyond technology. It is about ensuring that African languages remain visible, usable, and alive in the age of AI.
