WhatsThis?: Instantly understand screen content with AI. Snapshot, analyze, chat - all in one sleek, Linux-friendly app. Explore visuals effortlessly!
README for WhatsThis? - Version 2 Introduction Welcome to WhatsThis?, a snazzy, innovative application that simplifies the process of deciphering and discussing visual content on your screen. With just a snapshot, WhatsThis? uses the power of AI to analyze and describe any part of your screen. Whether it's an intriguing photo, a complex graph, or a snippet of text, WhatsThis? is your go-to tool for instant understanding.
Features Snapshot Analysis: Easily take a screenshot of a part of your screen and get an AI-powered description of its contents. Interactive Chat Interface: Engage in a dialog with the AI about the screenshot, asking further questions and gaining deeper insights. User-Friendly: Simple and intuitive GUI, tailored for ease of use. System Requirements Operating System: Tested on Linux. Dependencies: Ensure you have Python and the necessary libraries installed.
Installation To get started with WhatsThis?, follow these simple steps:
Clone the repository to your local machine.
git clone [repository-link] Navigate to the cloned directory.
Placing the OpenAI Key To use WhatsThis?, you need to place your OpenAI API key in the project folder for easy access. Simply create a text file named api_key.txt in the same directory as the WhatsThis? script. Paste your OpenAI API key into this file. The script is configured to automatically read the key from here, ensuring a smooth and secure operation of the application. This method keeps your key conveniently accessible while maintaining the simplicity of the setup process.
Usage WhatsThis? is designed for ease of use:
Launching the App: Navigate to the application directory and run the Python script.
python3 whatsthis.py Taking a Snapshot: Simply select the part of the screen you want to analyze. Interacting with AI: Use the chat interface to ask questions and get descriptions of the snapshot. Creating a Terminal Alias (For Linux Users)
For quick access, you can create an alias in your .bashrc or .bash_aliases file:
Open your .bashrc or .bash_aliases file in a text editor.
nano ~/.bashrc Add the following alias (modify the path to where the script is located):
alias whatsthis='python3 /path/to/whatsthis.py'
Reload your bash configuration: source ~/.bashrc
Important Notes Customizing Max Tokens WhatsThis? is designed to be flexible and adapt to various user needs. One of the key features you can customize is the max_tokens parameter in the API call. This parameter determines the length of the response generated by the AI. Adjusting max_tokens can be particularly useful depending on your specific requirements – whether you seek more detailed explanations or prefer concise summaries.
For those who require in-depth analysis or extended descriptions, increasing the max_tokens value will allow the AI to generate more comprehensive responses. Conversely, if you're looking for quick, to-the-point answers, reducing the max_tokens can provide shorter and more succinct responses. You can experiment with different values to find the balance that works best for your particular use case.
This customization can be particularly advantageous in diverse scenarios, ranging from academic research, where detailed explanations are crucial, to fast-paced environments where time is of the essence and brevity is key. By fine-tuning this setting, you can tailor WhatsThis? to be an even more effective tool for your visual analysis needs.
Path and Directory: Be mindful of the paths and directories specific to your system setup. Version: This is Version 2 of WhatsThis?, featuring enhanced AI integration and improved user experience.
Contributing Your contributions are welcome! Feel free to fork the repository and submit pull requests.
License Distributed under the MIT License. See LICENSE for more information.
Acknowledgments OpenAI for their incredible GPT-4 and image analysis capabilities. Enjoy exploring the world around you with WhatsThis? - Your AI-powered visual interpreter!