This project will involve several opportunities for students to participate in: (1) identification of cultural differences in communication involving how emotion is generated and recognized by speakers and listeners as well as (2) methods to discover a non-native English speaker's confusion or other indicators of distress in communicating with an American English speaker with no (or very little) understanding of the speaker's language and culture. This work will help students to discover the features useful in identifying differences in NLP and speech that are important for these projects and will provide experience in developing techniques for cleaning video data, extracting useful speech features from the data, and developing machine learning models to provide results. They will also learn how to identify cultural differences using speech and language models. Students will have access to Columbia Speech Lab machines and will collaborate with a number of other students working on the same general project with other CS faculty.
Lab: Columbia Speech Lab
Direct Supervisor: Julia Hirschberg
Position Dates: Summer 2022
Hours per Week: 20
Number of positions: 2
Qualifications: Computer Science
Eligibility: Freshman, Sophomore, Junior