In my SuperUROP I will be analyzing the distribution of phonetics across the world’s languages, applying Artificial Intelligence algorithms and techniques to develop phonetic models. We hope to use these models to develop a phonetic prior model that can guide computers as they try to learn the phonetic inventory of an unknown language. Based on what sounds are frequently used in common languages, we are able to use this as training data to predict what sort of sounds an entirely new language will most likely use. This project can help solve the problem of documentation of obscure foreign languages and also improve current Natural Language Processing techniques, voice recognition, and our understanding of common languages.

Last year I UROPed for the Media Lab’s Fluid Interfaces Group, doing Google Glass development and researching its applications. Over the past two summers I have worked for Amazon and Pebble, two companies that helped me learn and develop my skills. After taking several classes related to Artificial Intelligence, I really wanted to dive deeper into the field with a SuperUROP, which is how I discovered this group at CSAIL!