Available on npm

Voicci CLI

For Claude Code, OpenCode & AI Code Editors

AI-powered audiobook generation with natural language. Works with Claude Code, OpenCode, Cursor, Windsurf & more. Convert books and PDFs using XTTS v2. 100% local processing.

⚙️ Prerequisites (one-time setup)

Voicci generates speech locally using XTTS v2, so a small Python environment is needed alongside the npm CLI. Install these once, then you're set for life.

  1. Node.js 22+  ·  node -v  (older versions will not work)
  2. Python 3.10+  ·  python3 --version
  3. TTS + PyTorch  ·  pip3 install TTS torch  (on first audiobook, XTTS v2 weights ~2 GB download automatically)

After install, run voicci doctor — it verifies every prerequisite and prints a PASS/FAIL table.

📦 Quick Install

✨ Works with Claude Code, OpenCode, Cursor, Windsurf & more

Open your terminal and run:

npm install -g voicci

✅ Verify the install

Run this to confirm Node, Python, TTS and the editor skills are all wired up:

voicci doctor
🔁 Restart your AI editor after install so it picks up the new /voicci skill.

Where to run this command:

  • macOS: Open Terminal (Applications → Utilities → Terminal)
  • Linux: Open your Terminal application
  • Windows: experimental — WSL2 recommended; native PowerShell may hit PyTorch/TTS build issues
📦 View on npm

✅ After Installation - Usage with AI Editors

🎯 Claude Code (Best Experience)

Open Claude Code and type:

/voicci Lord of the Rings

Claude AI will search, download, and convert the book for you automatically!

🤖 OpenCode

Just tell the AI what you want:

"Find and convert 'Attention Is All You Need' paper to audiobook"

OpenCode will understand and run the voicci commands for you.

💬 Cursor / Windsurf / Other AI Editors

Use natural language:

"Convert this PDF to an audiobook"

"Find a book about machine learning"

"Summarize this paper for me"

Your AI assistant will intelligently use voicci commands to help you.

💡 Why Use AI Editors?

  • AI handles book search and selection intelligently
  • Automatic error troubleshooting and retries
  • Natural language - no need to remember commands
  • Progress monitoring and status updates
  • Smart summarization and content analysis

✨ Features

🎧
Audiobook Generation
Convert text and PDF files to natural-sounding audiobooks using XTTS v2 voice synthesis
🤖
AI Text Summarization
Condense long texts into concise summaries using AI-powered analysis
🔒
100% Local Processing
All processing happens on your machine. No data sent to external servers.
Fast & Efficient
Optimized queue system for processing multiple files concurrently
📖
Multiple Formats
Supports .txt, .pdf, and various text formats as input
🎨
Voice Customization
Use custom voice samples or built-in voices for narration

🚀 Usage

Convert Your Own Files

Convert text or PDF files that you own or have permission to use:

voicci mybook.pdf voicci document.txt --voice custom_voice.wav

Summarize Text

Generate an AI summary of a document:

voicci summary article.txt voicci summary research.pdf --output summary.txt

Options

voicci --help # Show all available commands voicci audiobook --help # Show audiobook-specific options voicci summary --help # Show summarization options

📋 Requirements

Voicci CLI requires:

  • Node.js v16 or higher
  • Python 3.8+ (for XTTS v2 voice synthesis)
  • 4GB RAM minimum (8GB recommended for faster processing)
  • Storage: ~2GB for voice models and dependencies

Platform Support

  • ✅ macOS (Intel & Apple Silicon)
  • ✅ Linux (Ubuntu, Debian, Fedora)
  • ✅ Windows (with WSL2 recommended)

🔗 Links & Support

📦
NPM Package
💻
GitHub Repository
📧
Contact

🎙️ Looking for Real-Time Voice to Text?

Check out our Voicci Mac App for instant voice transcription with a simple keyboard shortcut.

Try Voicci Mac App Free