10xChengTu

10xChengTu / input0

Public

Input0 — A macOS voice input tool: hold a hotkey to record, release to transcribe locally via STT, refine with LLM, and auto-paste into the active text field.

input0.com accessibility desktop-app llm local-first macos
14
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
Rust
AI Summary

Input 0 is a privacy-focused macOS app for voice-to-text that records on hotkey press, transcribes locally using GPU-accelerated AI models, refines output with an external language model, and auto-pastes into any application.

How It Works

1
📱 Download the App

Grab the free macOS app from GitHub releases and drag it into your Applications folder like any other app.

2
🔓 Allow Simple Permissions

Give it microphone access to hear you and accessibility to paste text anywhere on your Mac.

3
🔑 Link Your AI Helper

Enter a quick key from your AI service so it can polish your words into perfect text.

4
⬇️ Choose a Voice Listener

Pick and download a speech model that matches your language – it happens fast with a progress bar.

5
🎤 Just Speak Naturally

Hold Option+Space (or your custom key), talk as you normally would, release – and watch clean text appear in any app.

✨ Effortless Voice Typing

Your words turn into polished, ready-to-use text instantly pasted wherever you're typing, saving you endless keystrokes.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 14 to 14 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 input0?

Input0 is a macOS app built in Rust that lets you hold a hotkey to record voice, then release to transcribe it locally via STT engines like Whisper or SenseVoice on Apple Metal GPU, refine the output with an LLM, and auto-paste the polished text into the active input field. No window switching or cloud uploads needed—it's private, fast voice input for Slack, VS Code, browsers, or any app. Developers get seamless dictation without typing boilerplate.

Why is it gaining traction?

It stands out with on-device STT supporting 99+ languages across four engines, custom vocabulary for tech terms, and LLM refinement that strips fillers while structuring text. The hold-to-record/release-to-paste flow feels effortless, with ESC cancel, history comparison, and model auto-recommendations. Early adopters love the zero-friction integration into daily workflows over clunky browser extensions or paid services.

Who should use this?

Mac-based developers dictating code comments, commit messages, or Slack threads; technical writers refining meeting notes; or product folks brainstorming in WeChat. Ideal for frontend/backend engineers handling multilingual input or anyone tired of keyboard lag in active fields.

Verdict

Worth trying for Mac voice workflows—solid docs and bilingual README make onboarding easy despite 14 stars and 1.0% credibility signaling early maturity. Non-commercial CC BY-NC license limits enterprise use, but it's a promising Rust/Tauri gem for personal productivity.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.