giusmarci

giusmarci / openwhisp

Public

Voice to text, entirely on your machine. Hold Fn, speak, release — transcribed, polished, and pasted. No cloud, no account, no latency.

dictation electron local-ai macos ollama
12
1
100% credibility
Found Apr 12, 2026 at 14 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

OpenWhisp is a macOS desktop app that captures voice input via the Fn key, transcribes it locally, enhances the text with a local AI model, and pastes the result into the active text field.

How It Works

1
🔍 Discover OpenWhisp

You find OpenWhisp, a handy Mac app that turns your spoken words into neat, ready-to-use text without sending anything online.

2
📦 Get the helper ready

Download and start a free local AI helper app to make your words sound even better.

3
🚀 Launch the app

Follow simple steps to download and open OpenWhisp on your Mac.

4
🧙 Follow the setup guide

The app walks you through connecting the helper, grabbing voice and text tools, and allowing microphone and keyboard use.

5
🎤 Hold Fn and speak

Click in any text spot like an email or note, hold the Fn key, talk naturally, and let go to see your words appear polished and pasted.

6
✨ Pick your style

Choose between everyday chat or coding vibes, and how much polish you want, for perfect results every time.

✅ Dictate effortlessly

Now you speak your thoughts into emails, notes, or code super quickly and privately, feeling like magic.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 14 to 12 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 openwhisp?

OpenWhisp is a macOS desktop app built with TypeScript, Electron, and React that turns voice to text entirely on your machine—no cloud required. Hold the Fn key, speak naturally, and release to get transcribed text polished by a local LLM, then auto-pasted into your active app like a code editor or email. It solves the hassle of slow, privacy-invasive dictation tools by running Whisper for speech-to-text and Ollama for refinement, delivering instant, local copy voice to text.

Why is it gaining traction?

Unlike cloud-based options with latency and accounts, OpenWhisp keeps everything offline, auto-launches Ollama, and offers styles like "Vibe Coding" for dev-speak or Conversation for notes. A setup wizard handles permissions and model downloads (~10GB total), plus features like intent resolution for mid-sentence fixes and an audio-reactive overlay. Devs dig the zero-setup-after-first-run flow and direct paste into any text field.

Who should use this?

Mac-based developers dictating PR descriptions, Slack messages, or commit notes without typing. Frontend or backend engineers wanting quick, polished voice input for design docs or emails during pair programming. Anyone on Apple Silicon tired of cloud voice tools for sensitive work.

Verdict

Try it if local open whisper github alternatives appeal—solid weekend-built docs and MIT license make openwhisp review promising, despite 12 stars and 1.0% credibility score signaling early maturity. Download for dev machines; expect tweaks as it grows.

(178 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.