Emotion Recognition from Voice in the Wild

Abstract

Accurately recognizing emotionfrom voice is important in defenseapplications such as speakerprofiling and human-machineteaming, but is currently infeasible.We introduce a new, continuousspeech emotion recognitiondatabase, CMU-SER, and a set ofmicro-articulometry techniques thatcan capture finer nuances than thecurrent state of the art.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 2019
Accession Number
AD1118400

Entities

People

  • Oren Wright
  • Richard M. Stern
  • Rita Singh

Organizations

  • Carnegie Mellon University

Tags

DTIC Thesaurus Topics

  • Copyrights
  • Databases
  • Deep Learning
  • Department Of Defense
  • Dimensionality Reduction
  • Engineering
  • Governments
  • Guarantees
  • Learning
  • Machine Learning
  • Materials
  • Recognition
  • Signal Processing
  • Software Development
  • Universities

Fields of Study

  • Computer science

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Data Mining and Knowledge Discovery.
  • Systems Analysis and Design