What do Accuracy, Fluency, Read and Pace mean on a sayit result?

Every scored take in sayit breaks down into four bars — Accuracy, Fluency, Read, and Pace — and each one answers a different question about the same recording. They're kept separate on purpose: a take can be accurate but rushed, complete but hesitant, or fluent but unfinished, and folding all of that into one grade would hide exactly the information you need to know what to work on next.

30-second version: Accuracy is how close your phonemes were to the target sounds, averaged over the words you attempted. Fluency covers your speaking rate, pausing and filler-word density. Read is completeness — what fraction of the passage you actually attempted, only shown in read-aloud mode. Pace reports your words-per-minute and whether your timing ran rushed, dragging, or well-matched to the text. An overall score blends the first two, scaled by the third.

What does Accuracy actually measure?

The mean of your per-word pronunciation scores — how closely the sounds you produced matched the target phonemes, averaged across every word you were scored on. This is the number driven directly by the phoneme recognizer, and it's the same signal behind every per-word check or unclear verdict in your results. A word that scores low on accuracy is one where the sounds detected genuinely diverged from what the word required.

What's the difference between Fluency and Pace?

They're related but answer different questions. Fluency is the broader read on how smoothly your speech flowed — your overall speaking rate against a natural band, plus a penalty for excessive silence and for filler-word density. Pace is narrower and more mechanical: it reports your actual words-per-minute and checks the gap between every pair of words against the ideal for that spot in the sentence, then rolls up into a plain-language read — well-paced, rushing, over-pausing, or uneven. You can be fluent overall while pacing unevenly in a few specific spots, and the two bars can genuinely disagree with each other for exactly that reason.

What does "Read" mean, and why doesn't it always appear?

Read is completeness — the share of a known passage you actually attempted, as opposed to skipped or never reached. It only applies in read-aloud mode, where sayit knows the full target text in advance; there's nothing to measure completeness against in freeform speech, since there's no fixed passage length to compare to.

The four scores at a glance

How do these combine into one overall score?

Accuracy and fluency blend into an overall number, weighted toward accuracy, and in read-aloud mode that blend is then scaled down by how complete the read was — so an unfinished passage can't average out to a misleadingly high score just because the part you did read went well. Pace doesn't feed directly into the overall number; it's reported alongside as its own diagnostic, because pacing problems and accuracy problems call for genuinely different fixes.

Why does sayit keep these apart instead of giving one grade?

Because the fix for each is different. A low Accuracy score means practice the specific sounds that were flagged. A low Fluency or Pace score means slow down, breathe at the right points, or cut hesitation markers — a completely different kind of practice. A low Read score just means finish the passage next time. Collapsing all of that into a single number would tell you that something needs work without telling you which thing.

Try it

Read any passage in sayit and look at the four bars in your results together — if one is noticeably lower than the others, that's the one worth practicing next, and the individual articles on accuracy, fluency and pace go deeper into each.