Phase 1.5 · Articulation renderer v0

See the sound while you hear it

The renderer follows the canonical segment timeline from the accepted reference recording. Each pose comes from semantic articulation data.

Reference1963 ms
0.00s / 1.96s
Reference silencerest
Schematic target articulation. It is not a measurement of the learner's mouth.

Phase 1.8 · Your turn

Imitate the reference

Say I think this is a good idea. The recording is temporary. Technical quality is checked first; pronunciation assessment is a separate explicit step.

idle
Input level