Beacon listens to (or reads) a child's answers and turns them into a clear reading level and math level in minutes — no manual scoring, no spreadsheets, and no special training to run it.
For years, organizations have sent trained volunteers into schools with a clipboard and a stopwatch, sitting with one child at a time to check whether they can read a simple paragraph or solve a basic sum. This is how ASER, EGRA, and EGMA — three well-established, real assessment methods used around the world — actually work in the field.
Beacon does the same job, through an API. A child answers out loud into a microphone (or types an answer), and Beacon listens, transcribes, and scores it the way a trained evaluator would — instantly, consistently, at any scale.
Beacon's scoring is a solid starting point, not yet a certified instrument — it should be reviewed and tuned by education experts before it's used to make real decisions about a child's schooling.
Every organization's needs are different — Beacon offers three independent, real assessment methods instead of one-size-fits-all.
Adaptive — it gets easier or harder depending on how the child answers, finishing in just a handful of questions with one reading level and one math level. Best for a fast, good-enough snapshot.
The same fixed set of reading tasks for every child — naming letters, sounding out made-up words, reading a short story aloud, and answering questions about it. Timed and comparable across children.
The same fixed set of number tasks for every child — comparing numbers, spotting the missing number in a pattern, and solving simple sums against the clock.
Your app tells Beacon which child and which check to run.
They speak into a mic, or type — whichever fits the question.
Speech is transcribed and checked automatically, the way a human evaluator would.
A clear level or score comes back instantly — no manual grading.
Full endpoint reference, request/response schemas, and examples for all three schemes.