Wordle Solver is an innovative application designed to solve the New York Times Wordle puzzle efficiently. Leveraging advancements in computer vision and automation, the project combines OpenCV for accurate screen object recognition, Tkinter for an intuitive user interface, and PyAutoGUI for seamless interaction with the game interface. The system identifies the game's current state, calculates the optimal word guess, and inputs it automatically. Wordle Solver demonstrates how computer vision and automation techniques can work harmoniously to solve real-world challenges in recreational contexts.
A Tkinter-based GUI allows users to launch the application, configure settings, and observe real-time progress.
OpenCV is employed to detect and interpret game elements, such as the letter tiles, their positions, and color-coded feedback indicating correctness.
Frame-by-frame analysis ensures accuracy in identifying the game's current state after each guess.
PyAutoGUI interacts with the game by typing the calculated word guesses and simulating mouse clicks.
Automated keyboard inputs are synchronized with OpenCV's real-time feedback recognition.
A wordlist-based algorithm narrows down possibilities using color-coded feedback from the game.
The algorithm dynamically adjusts future guesses to minimize the number of attempts. I have fixed initial 4 words. These words don't contains any intersection of letters. In this way we are able to check 20 letters in 4 attempts
Robust exception handling ensures the system adapts to unexpected changes in the game's UI or errors in recognition.
Efficiency: The system successfully solves Wordle puzzles within the 6-guess limit in over 98% of test cases.
Speed: The solver completes each puzzle in an average of 30–40 seconds, including processing and automation time.
Accuracy: OpenCV achieves a recognition accuracy of over 95% for screen elements, even under varying lighting conditions.
User Experience: The interactive GUI makes the application accessible for users of varying technical expertise.
There are no models linked
There are no datasets linked
There are no datasets linked
There are no models linked
