Voice-first · AI-native · Africa-owned
An outbreak is heard long before it's confirmed.
OutbreakVoice turns a voice on a basic phone into a real-time early-warning signal — closing the 10-to-20-day gap between the first community deaths and the formal alarm.
No smartphone. No app. No internet. Just a voice, in the languages people speak.
The problem, in two numbers
The deadliest delay isn't diagnosis. It's detection.
In remote and cross-border settings, the disease is spoken of in villages long before it reaches the surveillance system. By the time a lab confirms a case, the chains are already spreading.
From the first community deaths and unusual symptoms to their arrival in the formal system.
From a community voice report to a verified, geo-located alert in the national system.
How it works
Five steps, from a spoken word to a saved life.
One integrated pipeline, built on the infrastructure that already exists in the most remote settings on earth: the human voice and a basic phone.
Ingest
Communities report symptoms and deaths by voice, SMS, or USSD — any phone, over 2G, in their own language.
Interpret
AI transcribes each report and fuses it with routine health-system signals into one early-warning score.
Cluster
Real-time space-time models separate genuine outbreak signatures from background noise and locate them.
Alert
Verified alerts flow into national DHIS2/IDSR systems and cascade to field teams — before lab confirmation.
Sharpen
Every field verification retrains the models, so accuracy improves with each outbreak.
Where it listens
From the village to the border crossing.
Because disease travels, OutbreakVoice reaches the community and the corridors between states — including the informal crossings that formal points of entry can't see.
The village
Health workers, traders, and traditional leaders as voice sentinels, plus community death investigations.
Airports
Travelers from high-risk origins self-report across the incubation window — sentinels, not quarantine.
Seaports
Onboard and port-health reporting, with a post-disembarkation follow-up window.
Transport corridors
Bus and boda unions and market associations as sentinels; USSD reporting at motor parks and transit hubs.
Cross-border
A shared dashboard so a signal on one side of a border is seen, with permission, on the other.
Built by the diaspora, owned by the continent
African data. African languages. African control.
Speaks the languages people speak
Swahili, Lingala, Hausa, Alur, Yoruba, Amharic, French, Arabic and more — reporting in a caller's own tongue, not an imported one.
Runs on what already exists
Built on the open DHIS2/IDSR standards African states already use, with no high-bandwidth requirement. It strengthens national systems rather than replacing them.
Data stays home
Per-country data residency and national ownership. The federation shares signals — not raw personal data — under African governance.
Keeps borders open
Verifiable early warning enables targeted action instead of trade-destroying blanket travel bans.
What we're building toward
The outcomes this platform is designed to deliver.
These are target outcomes — the benchmarks the program is designed and will be evaluated against, not results from the current prototype.
OutbreakVoice is in early development. The platform demonstrates the full workflow today — community reporting, cluster alerts, field response, and traveler monitoring — using simulated data. The statistical models, multilingual speech recognition at scale, and live national and cross-border integrations are being built. We're transparent about this because health security is earned, not claimed.
Partner with us
Give every African family an early-warning voice.
We're working with governments, Africa CDC, the diaspora, and global health partners to build this together. Tell us how you'd like to be part of it.
Thank you — your message is on its way.
We'll be in touch shortly. In the meantime, explore the platform.
Launch the platform ↗