MIT EECS | Hudson River Trading Undergraduate Research and Innovation Scholar
Analyzing Speech Signals for Clinical Diagnosis of Speech Impairments
- Natural Language and Speech Processing
Speech impairments are among the most common disabilities of childhood and can be very devastating. The limited methods of diagnosis are often complicated and inaccurate since children can have multiple speech impairments. This project focuses on using a speech-analysis system for clinical diagnosis of these impairments, analyzing modifications in distinct acoustic-phonetic aspects of features and phonemes in speech samples of children with various speech impairments. I will implement an alignment algorithm and perform analysis to find patterns of modification. I will further develop a system to improve the algorithm’ s capacity for diagnosis so doctors can address these problems more effectively. This will also have implications for automatic analysis of speech of typical speakers.
I am excited about participating in SuperUROP with a project that combines my majors of computer science and linguistics and that will help me make a real change in the world. I have taken courses in phonology, algorithms, and artificial intelligence, and I look forward to using all the skills and knowledge I acquired in those classes in the context of research. I also hope to grow more as a researcher during the program.