ClickUi is a Python-based, cross-platform AI assistant designed for local computer use, offering a unified interface for various AI models, speech recognition, and web scraping. It targets users seeking a powerful, integrated AI experience on their own hardware, providing both voice and chat interactions with features like file attachments and property lookups.

How It Works

The assistant leverages a modular architecture, integrating multiple AI models (OpenAI, Google, Ollama, Claude, Groq, OpenRouter) via API calls. Speech recognition is handled by Whisper, and text-to-speech by Kokoro. Web scraping and browser automation are managed using Playwright and Selenium. The GUI is built with PySide6, enabling a hotkey-toggleable interface for seamless interaction.

Quick Start & Requirements

• Installation: Recommended via Install.bat (Windows) or manual Conda environment setup (conda create --name click_ui --file conda_packages.txt, conda activate click_ui, pip install -r requirements.txt).
• Prerequisites: Python, Conda, Whisper, Kokoro (for voice mode). API keys for chosen AI models are required. ChromeDriver must match the installed Chrome version.
• Running: python clickui.py
• Demo: https://youtu.be/oH-A1hSdVKQ

Highlighted Details

• Supports local models via Ollama and various paid API providers.
• Features voice interaction with Whisper ASR and Kokoro TTS.
• Includes web scraping for Google searches and property lookups (Zillow/Redfin).
• File attachment support for .txt, .csv, .xlsx.

Maintenance & Community

The project is actively seeking collaborators. Feature requests and ideas can be submitted as checkboxes in the README. Pull requests are reviewed.

Licensing & Compatibility

The README does not explicitly state a license. Compatibility for commercial use or closed-source linking is not specified.

Limitations & Caveats

Voice mode dependencies (Whisper, Kokoro) must be installed and configured globally, otherwise runtime errors occur. Keyboard/mouse interaction is noted as having issues on Mac/Linux, with a planned fix using pynput. Some future features are marked as planned or in progress. Executable builds for cross-platform deployment are a future goal.