thainph

thainph / node-trans

Public

Real-time audio translation app

45
18
100% credibility
Found Mar 29, 2026 at 45 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
JavaScript
AI Summary

Node Trans is a desktop and web app for real-time audio transcription and translation from microphone or system sources, supporting cloud or offline modes with speaker identification and session history.

How It Works

1
🔍 Discover Node Trans

You find this handy app that turns spoken words into text and translates them live, perfect for meetings or videos.

2
📥 Get the app running

Download and launch it on your computer, opening a simple window ready to capture your voice or screen sounds.

3
🎤 Pick your audio and style

Choose if it listens to your microphone, computer sounds, or both, then select easy cloud mode or powerful offline setup.

4
Connect your translation helper
☁️
Cloud way

Enter a free service code once and start translating right away.

💻
Offline way

Follow simple steps to download voice tools that work completely on your machine.

5
▶️ Hit start and watch magic

Press play to begin live captions with speaker names, seeing words appear in real-time with translations.

6
📝 Review and share

Browse past sessions, rename speakers, or export notes, with an always-on-top window for quick glances.

✅ Effortless conversations

Now you understand every word in meetings, calls, or videos, bridging languages smoothly every time.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 45 to 45 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 node-trans?

Node-trans is a JavaScript real-time audio translation app that captures microphone, system audio, or both, then transcribes and translates speech on the fly using cloud Soniox or offline Whisper models. It handles speaker diarization to label "Speaker 1" vs. "Speaker 2," shows live partial results, and saves sessions with Markdown export—perfect for turning Zoom calls or podcasts into bilingual transcripts. Runs as a web app or Electron desktop build on macOS/Windows, with an always-on-top overlay for quick glances.

Why is it gaining traction?

It stands out with fully offline processing via Whisper.cpp and pyannote-audio, no API keys needed after setup, plus hardware acceleration on Apple Silicon for smooth real-time audio processing without cloud dependency. Dual-source capture (mic + system) and customizable translation via Ollama or LibreTranslate beat basic node translate tools, while the draggable overlay and session history make it practical for live use. Early adopters praise the bilingual UI and easy model downloads.

Who should use this?

Remote devs in multilingual teams capturing meetings, podcasters needing real-time audio to text with speaker labels, or content creators transcribing YouTube/Zoom without subscriptions. Ideal for Windows/macOS users with NVIDIA/Apple hardware wanting offline real-time audio translation extension alternatives to pricey cloud services.

Verdict

Worth a spin for offline real-time audio translation needs—solid docs and quickstart make setup straightforward despite 45 stars and 1.0% credibility score signaling early maturity. Test on your hardware first; lacks broad testing but delivers core value out of the box.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.