Statistical Speech Segmentation Lab

Stream Controls

Stream speed 2.5 syl/s

Boundary threshold τ 0.40

Live Statistics

TP(prev→cur) = —

0.0τ = 0.401.0

Syllables heard

Boundaries found

Segmentation accuracy

—

Syllables in language

12 / 12

How it works

The stream is built by concatenating 4 hidden three-syllable "words" back to back, with no pauses, stress cues or silence between them — exactly the Saffran, Aslin & Newport (1996) design that showed 8-month-old infants can segment such a stream after only ~2 minutes of listening.

TP(B | A) = count(A → B) / count(A)

within a word:  TP ≈ 1.0   (that syllable pair always co-occurs)
across a word boundary: TP ≈ 1/4 (the next word is one of 4, chosen ~uniformly)

As each new syllable arrives, the simulator updates a running bigram table and computes the transitional probability from the previous syllable to the current one using only statistics accumulated so far — an online learner, just like an infant's brain. Whenever TP drops below the threshold τ, it inserts a segmentation boundary (a gap in the conveyor).

Stream speed — how fast new syllables arrive.
Boundary threshold τ — how big a TP dip must be before the learner commits to a boundary; too high and it over-segments, too low and words run together.
New Language — reshuffles which syllables form which of the 4 hidden words and restarts learning from zero statistics.
Segmentation accuracy — the fraction of detected boundaries that land on a real word edge (a multiple of 3 syllables into the stream), scored against the true, hidden word grid.

This statistical-learning mechanism is one candidate explanation for how infants bootstrap word discovery before they know any actual words, and deficits in it are studied as a possible contributor to Specific Language Impairment and dyslexia.

Syllable block Correct boundary False boundary

⚠ Couldn't load the 3D engineThree.js failed to load from the CDN. Check your connection and reload.