Social personality evaluation based on prosodic and acoustic features

Yingnan Zhang, Jing Liu, Jin Hu, Xiang Xie, Shilei Huang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

6 Citations (Scopus)

Abstract

In recent decades, personality as a long term paralinguistic information has attracted more and more researchers. The main idea of the personality refers to the characteristics which acts as interactions between persons and the social occasions This paper proposes an approach for the automatic prediction of the Big-Five personality traits and 30 sub dimensions the listeners attribute to a speaker they don't know. The experiments are performed over a corpus of 1031 speech clips (337 identities in total) annotated not only Big-Five personality traits, but also all 30 sub-dimensions by using The Revised NEO Personality Inventory (NEO PI-R). The results show that it is possible to predict some particular subdimension with high accuracy (more than 75%) whether a person is perceived to be in the higher or lower part of the scales corresponding to each of the 30 sub dimensions, these sub dimensions give personality more accurate descriptions to lay the foundation for a more diversified personality classification.

Original languageEnglish
Title of host publicationProceedings of 2017 International Conference on Machine Learning and Soft Computing, ICMLSC 2017
PublisherAssociation for Computing Machinery
Pages214-218
Number of pages5
ISBN (Electronic)9781450348287
DOIs
Publication statusPublished - 13 Jan 2017
Event2017 International Conference on Machine Learning and Soft Computing, ICMLSC 2017 - Ho Chi Minh City, Viet Nam
Duration: 13 Jan 201716 Jan 2017

Publication series

NameACM International Conference Proceeding Series

Conference

Conference2017 International Conference on Machine Learning and Soft Computing, ICMLSC 2017
Country/TerritoryViet Nam
CityHo Chi Minh City
Period13/01/1716/01/17

Keywords

  • Classification
  • NEO-PI-R
  • Personality
  • Prosody and acoustics features

Fingerprint

Dive into the research topics of 'Social personality evaluation based on prosodic and acoustic features'. Together they form a unique fingerprint.

Cite this