Predictive Text: A Markov Model
    Generate text one word at a time using a simple Markov model trained on a chosen corpus.
    Markov Model — Text Prediction

    How Does a Machine Predict the Next Word?

    This tool uses a Markov model — one of the simplest forms of predictive text — to generate sentences one word at a time. It learns patterns from a corpus and, for each step, greedily selects the most probable next word given the current word.

    Step 1 — Choose a corpus

    Each option below is a pre-loaded historical speech excerpt. Try different speakers and compare how the corpus shapes the output.

    Or — Paste your own text
    Provide at least 100 characters.
    0 characters — need at least 100
    Step 2 — Build your prediction
    Select a corpus above to begin.