What is Vocoding?

Vocoding is a voice-first AI workspace — a desktop app that transforms spoken instructions into structured outputs using specialist agents, profiles, and context-aware optimization.

It is NOT just a transcription tool. Vocoding understands your project context, selects the right optimization profile and agent, and delivers structured prompts ready to paste.

The Core Flow

[Your Voice] → [Local Transcription] → [LLM Optimization] → [Structured Prompt] → [Clipboard]
     ↓                ↓                      ↓                     ↓
  Speak naturally   Whisper (100% local)   Cloud optional    Auto-copied, ready to paste

Time from voice to clipboard: < 5 seconds

Quick Start (5 Minutes)

1. Launch Vocoding

The app starts minimized to the menu bar (macOS) or system tray (Windows). Click the icon to open the main window, or use the global hotkey.

2. First-Run Setup

On first launch, Vocoding shows an onboarding wizard that guides you through:

Downloading a Whisper model for local transcription
Choosing your preferred operating mode
Configuring an LLM provider (optional)

3. Press the Global Hotkey

Platform	Default Hotkey
macOS	`Option + Space`
Windows	`Alt + Space`

4. Speak Your Idea

Talk naturally. For example:

"Create a React component for a user profile card that shows avatar, name, email, and a follow button. Use TypeScript and Tailwind CSS."

5. Get Your Result

Vocoding will:

Transcribe your voice locally (Whisper — 100% offline)
Optimize with LLM (if configured)
Copy the result to your clipboard automatically

6. Paste Anywhere

Press Cmd+V (macOS) or Ctrl+V (Windows) in your AI tool of choice.

Next Steps

Installation — System requirements and setup
First-Time Setup — Configure Whisper and LLM providers
The Three Modes — Transcribe, Optimize, and Chat