D.A.R.V.I.S.

Digital Assistant, Rather Very Intelligent System

A voice-activated AI assistant powered by Ollama Cloud + ElevenLabs TTS. Works on macOS, Android (Termux), and Linux.

Install (one command)

Open your terminal (or Termux on Android) and paste:

curl -sL https://raw.githubusercontent.com/Juanshep1/DARVIS/main/install.sh | bash

That's it. It installs everything, sets up API keys, and launches DARVIS.

After install, just type darvis anytime to start it again.

Features

• Voice input + output with ElevenLabs premium voices
• Ollama Cloud LLM brain (llama3.3, deepseek, qwen, etc.)
• Real-time web search + URL fetching
• File & folder creation, moving, copying
• Opens files/folders/URLs in system apps
• Safari browser control (macOS) — click links, read pages, navigate
• Native browser support (Android) — opens Chrome/Firefox/default
• Persistent settings (model + voice saved between sessions)
• /type and /listen modes for flexible input

Commands

Command	Action
/type	Pause mic, text-only input
/listen	Resume microphone listening
/voices	Interactive voice picker
/voice NAME	Quick switch voice (e.g. /voice rachel)
/help	Show all commands
goodbye	Exit

Safari Control (macOS only)

Enable: Safari > Settings > Advanced > Show features for web developers > Develop > Allow JavaScript from Apple Events

Then ask DARVIS to click links, read pages, navigate, scroll, type into forms, etc.

Platform Support

Feature	macOS	Android (Termux)	Linux
ElevenLabs TTS	afplay	termux-media-player	mpv/ffplay
Fallback TTS	say	termux-tts-speak	espeak
Voice input	PyAudio + Google	termux-speech-to-text	PyAudio + Google
Open files	open	termux-open	xdg-open
Browser	Safari control	Native browser	Default browser