Morse Code Decoder

During the summer of 2020 I worked at the Tufts Center for Engineering Education and Outreach (CEEO), where I was placed on the AI team with the goal of creating projects for the LEGO SPIKE Prime that incorporated AI algorithms. My first project was a Morse code decoder.

The decoder uses the k-nearest neighbor algorithm to categorize button presses of different lengths as either dots or dashes. By recognizing the timing intervals between presses, the program can translate a sequence of Morse code input into its component English letters, words, and sentences.

Short button press classified as a dot Long button press classified as a dash

Left: a short press, classified as a dot. Right: a long press, classified as a dash.

To see the Morse Code Decoder in action please check out the video below:

← Back to projects