Capturing, Representing, and Interacting with Laughter
summary
A system extracts features from multi-hour home recordings (about three hours per participant, 20 participants) and classifies short clips with a deep neural network into laugh versus non-laugh; designers build tangible outputs: a fragile perfume bottle that releases a single laugh, a jar that releases collections with flickering lights, and hand-cut chocolate towers encoding quantity and laugh quality. Participants prefer automated capture (11 of 16) but raise privacy, selective preservation, and context trade-offs; 1-hour review sessions reveal tangibility shapes meaning.
AI-written summary · about
chapters (7)
- 0:00Introduction to laughter project
- 1:13Capturing laughter with ML
- 5:26Designing laughter containers
- 9:01Chocolate representations study
- 13:41Privacy and tangibility of laughter
- 15:51Future work and privacy issues
- 17:06Q&A session
related talks
33mWhen Stand-up Comedy Meets Machine Learning
33mWhat if a Robot Could Laugh at Your Jokes? (Laughter Prediction with Natural Language Processing)
59mTechnical Workshop: Sarcastically Speaking: Unlocking Multi-modal Sentiment Analysis With NLP and Facial Expressions
25mLooking at Sound
18mSketching CuddleBits: Coupled Prototyping of Body and Behaviour for an Affective Robot Pet
02mMood Video
15mRecognizing Emotional Expression in Game Streams
22mSeeing, Sensing and Recognizing Laban Movement Qualities