The Word That Started Everything
In Papua New Guinea — home to the world's greatest linguistic diversity — the word wantok carries enormous weight. Literally from Tok Pisin: "one talk" — someone who speaks your language. But in practice, it means so much more.
Your wantok is your person. The one you call when you're in trouble. The one who has your back across any distance. In a country where you might share a market with someone who speaks an entirely different language, finding your wantok is finding family.
We believe that feeling — that deep relief of being truly understood — is something every person deserves in every meeting.
The Problem We're Solving
Today, 60% of international business meetings have at least one participant who is not communicating in their strongest language. That person is processing slower, contributing less, missing nuance. The cost is invisible but enormous.
Human interpreters are extraordinary — and priced for governments and Fortune 500 companies. The rest of the world's meetings happen in English, whether or not everyone in the room speaks English.
We are changing that. With AI inference from NVIDIA NIM at effectively zero cost, we can offer professional-grade interpretation to any team, at any price point.
The DN Ecosystem Connection
Wantok is a product of the Depollute Now! ecosystem — a community working at the intersection of technology and environmental restoration.
Every subscription contributes 10% of revenue to plastic removal and reforestation programs. Every meeting interpreted is a small act of both connection and healing. We track it all, transparently, on our live impact counter.
When the Arbitrum Orbit L3 infrastructure is ready, users will receive DNOW tokens as tangible proof of their environmental contribution — a blockchain record of the world they helped build.
The Technical Miracle
What makes Wantok possible right now — not five years from now — is the convergence of three things:
- NVIDIA NIM free tier — models like qwen3-235b-a22b at zero inference cost, with 35 RPM per model and 10+ models available
- Zoom RTMS — real-time transcript delivery from Zoom meetings, with built-in ASR at zero additional cost
- Payload CMS v3 — a complete backend framework that eliminates months of infrastructure work
Total AI inference cost for the MVP: $0. Total hosting cost: $4.75/month on Contabo. This is what allows us to offer audio interpretation in the free tier — something no competitor does.