This project investigates acoustic cues in human speech – the final physically measurable signals before abstraction into phonemes and words. Because natural speech frequently differs from its canonical form through modifications such as deletions and insertions, analyzing them can reveal how speakers plan and execute speech production. LEXI aims both to understand these patterns scientifically
and to develop tools for their automatic recognition, with important potential clinical applications: early evidence suggests that children with language-related developmental disorders such as dyslexia or ASD may produce distinctive patterns of acoustic modifications that could serve as biomarkers for early detection.

I have been participating in this UROP for almost a year, and it’s been an incredible opportunity to combine my linguistics interests with hands-on computational work and data analysis. This SuperUROP would allow me to gain more experience with writing and presenting scientific results while working on the project I already love.