← All models · MMS forced alignment
use case · wild
The most personal alignment target there is: you. Record yourself saying a sentence, type what you said, and the model times every word of your own voice. This is the engine behind pronunciation apps and reading-coach tools — running entirely in your browser, so your voice never leaves your device.
CC-BY-NC-4.0 — non-commercial weights. For personal/educational use; see the main page for full license + attribution.
This rung records your own voice; the bundled JFK sample used on the other rungs is an inaugural address excerpt (20 January 1961), spoken by John F. Kennedy — public domain, U.S. government work. JFK audio source and attribution record.
Suggested sentence: "The quick brown fox jumps over the lazy dog." — say it slowly and clearly, then type exactly what you said below.
your words, timed — click to hear yourself
The aligner is honest: words that don't match the audio (background noise, a stumble, a changed word) show up in the matched-char ratio. Try reading a sentence from a book or a song lyric you know by heart.