Concept
This project is a voice-controlled Arduino car that allows a user to control a moving robot using simple spoken commands. Instead of buttons, remotes, or apps, the user speaks commands such as “forward,” “turn left,” “reverse,”or “stop.” The car listens and responds in real time.
The system uses the Web Speech API (built into modern browsers) to recognize speech. A fuzzy matching algorithmhelps the system understand different ways of saying the same thing. For example, if the user says “go forward,” “move forward,” or simply “forward,” the system will run the same command.
The computer communicates with the Arduino through serial communication (USB cable). Once the browser identifies the command, it sends a single character (f, b, l, r, s) to the car. The Arduino then interprets the command and drives the motors.
This creates a smooth and natural interaction where your voice becomes the controller.
Project Images
Schematics
User Testing
First User: User 1
Second User : User 2
IMPLEMENTATION
The implementation is designed to feel natural, simple, and automatic. When the user opens the web page, the system asks for microphone access and immediately begins listening for voice commands once permission is given. It listens continuously, and a microphone icon provides live feedback by turning into a green pulsing circle when actively listening and a gray circle when idle. When the user speaks, the microphone captures the audio, the Web Speech API converts it into text, and a fuzzy matching algorithm chooses the closest valid command, even if the person says it in a different way. The interface then shows what the system heard, which command was selected, whether the Arduino is connected, and the current auto-mode status, helping the user understand what is happening at all times. Once a command is recognized, the browser sends it through USB to the Arduino, which drives the motors using the Motor Shield so the car can move, turn, stop, or reverse immediately.
Parts I’m Proud of
I am especially proud of how smart and flexible the command recognition system is, because the fuzzy matching allows the car to understand many different ways of saying the same command, making the interaction feel natural. I also like that the user experience is very smooth—the system starts listening automatically, works without any extra buttons, and gives clear visual feedback showing what it heard and what command it chose. The error handling is strong as well, so even if the microphone fails, the Arduino disconnects, or a command is unclear, the system continues running and informs the user about what went wrong. The real-time indicators for listening status, recognized text, executed commands, and connection state make the experience easy to follow. I am also proud of the continuous listening design, which removes the need to press a talk button and makes the car react instantly to the user’s voice. Finally, the code is clean and uses modern web APIs and smart algorithms, giving the project a professional and reliable feel.
AI Assistance. : This project used AI tools mainly to support coding, hardware wiring, testing, and final implementation were completed by me. AI was used to help clarify technical concepts. It assisted in explaining algorithms such as fuzzy matching and Levenshtein distance, and in organizing the projects. Any AI-generated text was reviewed, edited, and adapted to fit the actual project work. The creative decisions, programming logic, user testing, debugging, and system design were all done by me, with AI acting only as a supportive tool for communication and explanation.