One person. One waveform.
XLNTalk is a one-person project, built and maintained from a small studio in Västerås, Sweden. No VCs, no team, no roadmap politics — just one person obsessing over the way voices travel.
Voice is the original interface. The software should disappear behind it.
I started XLNTalk in 2023 after spending too many hours on calls that sounded like calls. Compression artifacts, robotic latency, plastic timbre. The technology to do better had been sitting around for a decade — Opus, edge networks, real audio in browsers. The problem wasn’t the tools. It was that nobody was treating voice as the primary product.
So I built XLNTalk. Studio-grade fidelity, 12ms latency, end-to-end encryption — completely free, forever. No team to feed, no investors to please, no upsell engineered into the product.
Signal over noise
I obsess over the smallest details of audio: room tone, transient response, latency variance under 2ms. Everything else is decoration.
Privacy is a feature
Your conversations are encrypted end-to-end. I’ll never train AI on them, never sell metadata, and never will. This is non-negotiable.
Less, but better
I’d rather ship one feature that disappears into the experience than ten that beg for attention. Software should feel like absence.
Humans answer
Every email is read by me. Most are answered within a few hours. There is no chatbot, no ticketing portal — just a person, in Sweden.
A small desk in
Västerås, Sweden.
XLNTalk is built between coffee runs at a desk facing Lake Mälaren. Releases ship on Stockholm time. Support emails get answered before the lake freezes over.
- Västerås, Sweden · 60km west of Stockholm
- Solo dev · WebRTC · Rust · TypeScript
- Powered by espresso and Lake Mälaren wind
- hello@xlntalk.com
“I make XLNTalk because the voice apps I had to use were building louder, not clearer. So I sat down and started building the one I wished existed.”
XLN_LON 16.5448°E