Computational Modeling of Emotions and Affect in Social-Cultural Interaction
Abstract
The objective of this proposal was to develop computational models for emotion recognition in speech and study various impacting factors including social, cultural, and language effect on such models. Accomplishments in the project are the following. First, emotion recognition performance was improved upon the state-of-the-art. Different methods were developed to improve model performance, including employing sub-sentence units, advanced feature transform, deep learning, and more features from acoustic and textual information sources. Second, a cross-lingual study was performed that shed light on how human perception and automatic recognition of emotion differs and how performance in cross-lingual setups varies. This project supported several students, leading partly to one Ph.D dissertation
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 02, 2013
- Accession Number
- ADA591829
Entities
People
- Yang Liu
Organizations
- University of Texas at Dallas