Dataset provider for training, aligning, and evaluating AI models on African languages and industry domains. Email: [email protected]
Africa
Joined March 2025
- Tweets97
- Following6
- Followers128
- Likes22
Pinned Tweet
As we celebrate Tonative's first anniversary, we're excited to introduce TonalKeys - our Chrome extension designed to make typing African language characters and diacritics simpler.
Language should never be a barrier to creating, communicating, or preserving our stories online. TonalKeys is one more step toward making African languages easier to use in the digital world.
How to use:
1. Install TonalKeys on your desktop Chrome browser:
chromewebstore.google.com/de…
2. Once installed, simply toggle it ON whenever you need to write on any web page, and select your language of choice.
3. A floating keyboard will appear on your page, whether it's a Google Doc, an email, or anything else. If it doesn't appear right away, just refresh the page.
4. Whenever you need a special character, click the corresponding key on the floating keyboard. That's it.
We'd love to hear any feedback you have as we continue to improve it.
Thank you for being part of our first year. 🎉
#TonativeTurnsOne #TonalKeys #AfricanLanguageTechnology
Building AI models for African languages means doing a lot of things in cycles:
Idea formulation → Data curation → Model building → Model evaluation → Deployment → Monitoring → Marketing → and back again.
For most teams working on African languages, one of the stages that eat up the most time is data curation and this applies to training, post-training or evaluation datasets.
Due to licensing issues, you can't use any data. Often time, you have to create budgets, find curators yourself, train them, write the annotation guidelines, supervise, validate, and clean. Then you repeat this process until the quality is good enough for use.
But you don't have to bear this workload. At Tonative, we take data curation off your plate, so your team can put their energy in the right place, which is building, shipping and growing your product.
Book a free call with us to learn more on how we can support you: lnkd.in/egQ7U3Rm
In a new country but lost in conversations?
Our team travels a lot and their story tells a similar experience of how difficult it is to navigate conversations with people you do not share the same language with.
That exact problem has led us into our first product called Volang. It is a real-time translation app for natural, face-to-face and online conversations. It currently supports 45 languages, including Amharic, Swahili, Luo, Yoruba and Hausa.
We're about to start beta testing, and we want people who have experienced this problem before. You would get free access to @volang_app throughout the testing phase.
Join the waitlist here: volang.app/
Building Africa's Human Data Infrastructure for Sustainable AI.
At Deep Learning Indaba 2026, we brought together AI researchers, data collectors, and members of the NLP community to explore what it takes to build stronger foundations for African language data curators.
From conversations around community-led data curation, annotation tools, data infrastructure, and governance to hands-on discussions around the people and processes behind African language datasets, the workshop created space for meaningful exchange.
We were also proud to present our Paper and poster:
“Directional Hallucinations in African Neural Machine Translation: How Source Language Typology Shapes Translation Faithfulness.”
authored by Alfred Kondoro, Okechukwu God'spraise, CHIKEZIE OFODILE and Verrah Otiende, PhD.
The conversations reinforced an important idea: building better African AI requires not only better models, but also better data, stronger infrastructure, responsible practices, and empowered communities.
Thank you to everyone who participated and helped make the workshop meaningful. We appreciate the presence of Elias Dogbatsey, Robbie Phillips, Ousseynou MBAYE at the booth
#Tonative #DeepLearningIndaba #AfricanAI #AfricanLanguages #NLP #DataCuration #ResponsibleAI
From conversations to connections. 🤝
We had our team at the Deep Learning Indaba last week at Pan-Atlantic University, and during the community booth session, they met with researchers, founders, partners and innovators in the African AI space.
We discussed our work around context dataset, and listened to new perspectives. We appreciate the presence of Flora Oladipupo, Oluwafemi Azeez, Dr Sam Oonge, Dr Lilian Wanzare, Foutse Yuehgoh, Isabella Nascimento and others at the booth
Thank you to everyone who stopped by, asked questions, shared ideas and connected with us.
Until the next one. 🌍
Sharon Ibejih, Cynthia Amol, Chinenye Anikwenze, Okechukwu God'spraise, Eric Jagwara, Joy Olusanya, Hope, Terry, Ifeoma Okoh and Chikezie Amalagu
#Tonative #DeepLearningIndaba #AfricanAI #AfricanLanguages #AIForAfrica
Applications are now open for the Tonative Data Academy, a hands-on program designed to equip you with the skills to collect, validate, annotate and govern African language datasets for AI.
This course is open to linguists, researchers, developers or people who are interested in improving their language's representation in AI systems.
📅 Application deadline: August 30
💻 Cohort starts: September 7
💰Course Fee: $45
📍 Apply now: tonative.org/services/data-a…
At the end of the program, you will earn a certificate, get a portfolio dataset, collaborate on a research, and have access to join the team of data curators at Tonative.
#Tonative #TonativeDataAcademy #AfricaNLP #DataCuration
Heading to Deep Learning Indaba next week? So are we. 👋
Come find us as we share our research, host a workshop on building Africa's human data infrastructure, and meeting the amazing AI community.
Don't forget to stop by our booth on Wednesday, we can't wait to meet you!
#Tonative #DeepLearningIndaba #AfricanLanguages
Tonative retweeted
Should data curators play a role in licensing decisions?
It's a simple question, but one with significant implications for the future of African languages.
As we build datasets that power speech technologies, translation systems, and language models, questions around ownership, licensing, ethics, and governance become just as important as the data itself
At our Third breakout session during the Tonative × @DeepIndaba 2026 Workshop, we'll explore the policies, ethics, and governance frameworks needed to build responsible African language datasets.
Breakout Session 03: Data Governance for African Datasets
Join our moderators:
🔹 Cynthia Amol (@JayneAmol) – Co-founder & Head of Data, Tonative
🔹 Faisal Muhammad Adam – Hausa Language Lead, Tonative
Together, they'll lead discussions on:
• Data Licensing Awareness & Frameworks: Should data curators play a role in licensing decisions, and what licensing models best support African language datasets?
• Navigating Cross-Regional Ethics Gaps: How do we responsibly build datasets across different countries and communities with varying ethical standards?
These conversations are essential for anyone interested in creating AI systems that are not only innovative, but also ethical, sustainable, and community-centered.
📅 Date: August 6, 2026
#Tonative #DeepLearningIndaba #DataGovernance
Tonative retweeted
Looking forward to the session
After our keynote session, attendees will choose from focused breakout discussions.
Breakout Session 02: Annotation Tools & Data Infrastructure
Moderated by:
• God'spraise Okechukwu (@Prazehub)
• Chinenye Anikwenze (@kodervine)
We'll explore:
🔹 Localized annotation tools (open-source vs. proprietary)
🔹 The infrastructure needed to store, manage, and scale African language datasets
Afterward, we'll reconvene to share key insights across all breakout sessions.
📅 Aug. 6, 2026
#Tonative #DeepLearningIndaba #DataAnnotation
Should data curators play a role in licensing decisions?
It's a simple question, but one with significant implications for the future of African languages.
As we build datasets that power speech technologies, translation systems, and language models, questions around ownership, licensing, ethics, and governance become just as important as the data itself
At our Third breakout session during the Tonative × @DeepIndaba 2026 Workshop, we'll explore the policies, ethics, and governance frameworks needed to build responsible African language datasets.
Breakout Session 03: Data Governance for African Datasets
Join our moderators:
🔹 Cynthia Amol (@JayneAmol) – Co-founder & Head of Data, Tonative
🔹 Faisal Muhammad Adam – Hausa Language Lead, Tonative
Together, they'll lead discussions on:
• Data Licensing Awareness & Frameworks: Should data curators play a role in licensing decisions, and what licensing models best support African language datasets?
• Navigating Cross-Regional Ethics Gaps: How do we responsibly build datasets across different countries and communities with varying ethical standards?
These conversations are essential for anyone interested in creating AI systems that are not only innovative, but also ethical, sustainable, and community-centered.
📅 Date: August 6, 2026
#Tonative #DeepLearningIndaba #DataGovernance
After our keynote session, attendees will choose from focused breakout discussions.
Breakout Session 02: Annotation Tools & Data Infrastructure
Moderated by:
• God'spraise Okechukwu (@Prazehub)
• Chinenye Anikwenze (@kodervine)
We'll explore:
🔹 Localized annotation tools (open-source vs. proprietary)
🔹 The infrastructure needed to store, manage, and scale African language datasets
Afterward, we'll reconvene to share key insights across all breakout sessions.
📅 Aug. 6, 2026
#Tonative #DeepLearningIndaba #DataAnnotation
After our keynote with Dr. Lilian Wanzare, attendees will choose from focused breakout sessions @DeepIndaba .
Breakout Session 01: Capacity Building for Data Curation
Moderated by:
• Sharon Ibejih (@sharonibejih)
• Joy Olusanya (@Naomi_Joy___ )
We'll discuss:
🔹 Pathways into data curation
🔹 Data annotation as a sustainable career
Then we'll reconvene to share insights and shape the conversation on Africa's AI data ecosystem.
📅 Aug. 6, 2026
#Tonative #DeepLearningIndaba #DataAnnotation
We're excited to announce our Guest speaker for the Tonative × Deep Learning Indaba @DeepIndaba 2026 Workshop.
Please join us in welcoming:
Dr. Lilian Wanzare - Co-founder of KenCorpus and Research Lead at the Maseno Centre for Applied Artificial Intelligence (MCAAI), with a decade of experience curating datasets across Kenyan languages including Dholuo, Kikuyu, Kalenjin, Maasai, Somali and Kenyan Sign Language (KSL). As the Principal Investigator for KenCorpus, African Next Voices - Kenya and AI4KSL, she led the collection of some of the largest speech, text and sign language datasets for Kenyan languages, working directly with language communities throughout. Her research centres the empowerment of human data infrastructure as foundational to the AI development pipeline.
In this session, Dr. Wanzare will share insights on the capacity building for African language data curators, the infrastructure needed for a scalable curation pipeline and how to govern these datasets.
Our workshop is open to anyone attending the Deep Learning Inbada Conference this year, and is interested in the data layer of language technologies.
Kindly signify interest via the DLI registration portal:
baobab.deeplearningindaba.co…
📅 Date: August 6, 2026
#Tonative #DeepLearningIndaba #DataAnnotation
Our paper has been accepted at Deep Learning Indaba 2026! 🎉
We're excited to announce that our research paper, "Directional Hallucinations in African Neural Machine Translation: How Source Language Typology Shapes Translation Faithfulness Across Six African Languages," has been accepted for presentation at Deep Learning Indaba 2026.
This work examines how translation quality changes depending on the source language and shows that many translation errors in African languages are driven more by data availability than by the languages themselves. Our findings demonstrate that improving African AI requires greater investment in high-quality, community-driven language datasets not simply larger models.
The research evaluates over 36,000 translations across six African languages using leading multilingual machine translation systems, contributing new insights into building more faithful and inclusive AI for Africa.
Congratulations to our authors. We look forward to sharing our work at Deep Learning Indaba 2026.
#DeepLearningIndaba2026 #AIForAfrica #AfricanLanguages #MachineTranslation #NLP #ResponsibleAI #Research #Tonative
Celebrating the people behind our first year. 💚🧡
As Tonative marks one year, we're taking a moment to recognize the incredible community leads who have helped shape our journey.
From coordinating language communities and leading research to developing the software tools needed for our work, to communications and community engagement, each person has contributed to every datasets we have built and the stories we tell today.
Thank you so much for all the contributions you have led at Tonative. Looking forward to our next chapter.
Happy Anniversary, Team! 🎉
We are incredibly proud to have been a data curation partner for @ml_collective by providing and validating the datasets that fuel their research.
Huge congratulations to the entire MLC Africa team for consistently putting African research on the global stage. ✨
Looking forward to @DeepIndaba 2026.
Community Showcase: @ml_collective Africa 🔥
Building world-class AI research from Africa, for the world!
ML Collective Africa is a pan-African research community empowering the next generation of African AI researchers through mentorship, paper reading groups, and collaborative projects. From publishing at top-tier venues like NeurIPS, ICML, ACL, EMNLP, AAAI, and MICCAI, to securing prestigious global fellowships, their members are making remarkable contributions to the global AI landscape.
Their impact goes far beyond peer review. Through active grassroots fundraising, they have continuously supported members to attend multiple Deep Learning Indaba conferences while fostering deep collaborations with vital ecosystem partners like @MasakhaneNLP and @tonative_org . They are proving that Africa is a key creator and leader in the future of intelligence.
Watch their community showcase to see how collaboration, curiosity, and research excellence are shaping the continent's tech ecosystem.
Join the conversation from anywhere!
While in-person registrations for Lagos are closed, you can still experience the magic, keynotes, and deep learning sessions of the Indaba online.
💻 Virtual Indaba: Registrations are open to join #DLI2026 virtually! Secure your pass here 👉 deeplearningindaba.com/2026/…
#DLI2026 #Indaba2026
