Launched this week
NexTalk

NexTalk

The missing voice input for Linux. Beautiful Private Offline

83 followers

A modern voice input tool built exclusively for Linux. Featuring a transparent "capsule" UI, 100% offline inference (Sherpa-onnx), and native Fcitx5 integration. Sub-20ms latency.
NexTalk gallery image
NexTalk gallery image
NexTalk gallery image
NexTalk gallery image
NexTalk gallery image
NexTalk gallery image
NexTalk gallery image
NexTalk gallery image
Free
Launch Team
Migma AI
Migma AI
Lovable for Email
Promoted

What do you think? …

Decker
Maker
📌
Hi PH community! 👋 I'm Decker, the maker of NexTalk. For a long time, as a Linux user, I felt a bit... forgotten. While my friends on macOS and Windows enjoyed sleek, integrated voice dictation, our options on Linux were mostly a collection of messy scripts, outdated UIs, or clunky clipboard simulators that broke half the time. I kept asking myself: "Is a beautiful, high-quality voice experience really a privilege exclusive to Mac and Windows?" I decided the answer was "No." So I built NexTalk. My goal was simple: Create a voice input tool that feels like a natural extension of the Linux desktop—Invisible but Powerful. What makes NexTalk different: 🔒 100% Offline & Private: Your voice never leaves your machine. We use Sherpa-onnx for local inference because your privacy isn't a trade-off for convenience. 🎨 Aesthetics First: A minimalist, transparent capsule UI built with Flutter. No borders, no clutter—just a breathing light that listens. ⚡ Lightning Fast: Sub-20ms latency. It feels like the words are flowing directly from your mind to the screen. 🐧 Linux-Native: No more hacks. We integrated directly with Fcitx5 via Unix Sockets, ensuring perfect compatibility with Wayland and your favorite apps. NexTalk is my love letter to the Linux community. It's open-source and built for those who refuse to settle for "good enough." I’d love to hear your thoughts, bug reports, or feature ideas. Let's make Linux productivity better, together! 🚀
Decker

Hi everyone! Thanks so much for the incredible support during our launch week. We’ve been listening closely to your feedback, and I’m excited to share a fresh update for NexTalk.

One of the top requests was better control over audio sources. We’ve just released a new version that supports Manual Audio Input Switching. 🎙️

  • What’s new: If you have multiple microphones (e.g., a built-in mic and a USB headset), you can now manually select your preferred device directly in the settings.

We’re committed to making NexTalk the best voice tool for the Linux desktop. Update your build and let us know what you think! 🐧✨