phuc-nt

Real-time speech translation for macOS — open source, no tracking, direct Soniox API

44
12
100% credibility
Found Mar 14, 2026 at 21 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
JavaScript
AI Summary

A desktop application that captures audio from your system or microphone, transcribes and translates speech in real-time, and displays it in an overlay window using a third-party service.

How It Works

1
🔍 Discover My Translator

You find this cool app on GitHub that listens to speech from videos, meetings, or your microphone and translates it live into your language.

2
📥 Download and Install

Pick the right file for your Mac or Windows computer, open it up, and move the app to your programs folder to get started.

3
🔒 Grant Sound Permissions

The first time you launch it, allow the app to hear your computer's audio or microphone so it can capture what you're playing or saying.

4
🌐 Connect Translation Service

Visit a speech-to-text website, make a free account, grab your special access code, and paste it into the app's easy settings panel.

5
🗣️ Choose Languages

Select the language of the speech you want to translate from and the one you prefer to read it in, then save your choices.

6
▶️ Hit Play and Watch Live!

Press the big play button, start any audio like a YouTube video or conversation, and see smooth real-time translations pop up in a handy floating window.

7
💾 Keep Your Translations

Everything you translate gets saved automatically as simple text files on your computer, ready to read anytime.

🎉 Effortless Understanding

Now you can enjoy podcasts, meetings, or foreign videos fully translated into your language, feeling connected and in the loop.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 21 to 44 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is my-translator?

My-translator delivers real-time speech translation for macOS and Windows, grabbing system audio from YouTube, meetings, or podcasts—or direct mic input—and piping it to the Soniox API for transcription and translation across 70+ languages. It pops up a dark, always-on-top overlay with instant results, speaker labels, and local transcript saves, solving the hassle of pausing content to copy-paste for "my translator AI translate" or fixing "my translate not working in Spanish." JavaScript frontend with Tauri backend keeps it snappy and privacy-focused, no servers involved.

Why is it gaining traction?

No tracking or accounts—just paste your Soniox key (~$0.12/hour) for direct WebSocket streaming, beating browser tabs or paid apps with multi-speaker detection and auto-session resets every 3 minutes. Users rave about overlay tweaks for font, opacity, and lines, plus keyboard shortcuts for switching mic/system modes. It's a real-time speech recognition gem on GitHub, ideal for "my translator Chinese drama" or "K-drama" binges, "my translator image" workflows, or "real-time API GitHub" experiments.

Who should use this?

DevOps engineers transcribing global incident calls, indie creators subtitling "my translator drama" clips or "forces love" K-dramas, or linguists checking "my translator meaning in Tamil." Suits remote workers needing live translation for non-English standups or "real-time speech recognition" demos.

Verdict

Solid for early adopters despite 19 stars and 1.0% credibility—great docs and releases, but unsigned binaries need terminal tweaks, and Windows support is fresh. Fork it for real-time projects GitHub if Soniox fits your stack.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.