← escprojects/emoticode
04

Emoticode

ios · research · lead developer / researcher · shipped · — views

Undergraduate research turned app: emotional-intelligence training as a covert-operations game, where progress depends on actually expressing the emotion a mission calls for.

Face, voice and text are read together — TFLite face detection, Emo-AffectNet for facial emotion, Whisper for transcription, TweetNLP for sentiment — so the app can catch disagreement between what you say and how you say it. Presented at the MDC Undergraduate Research Symposium and the Florida Undergraduate Research Conference.

stack
Swift · Emo-AffectNet · Whisper · TweetNLP
role
Lead developer / researcher
links
not public yet
main screen
main screen
success case
success case
modality select
modality select
← NORMALprojects/emoticode4/5 · utf-8