The acoustic fingerprint of a charismatic voice - Initial evidence from correlations between long-term spectral features and listener ratings

Oliver Niebuhr, Radek Skarnitzl, Lea Tylečková

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

508 Downloads (Pure)

Abstract

What makes a charismatic speaker? The present study extends this question into the prosodic dimension of voice quality. We analyzed various F0, LTAS and LTF long-term spectral characteristics from 12 L2 speakers of English who were recorded while giving entrepreneurial speeches. The results of the acoustic analysis were correlated with indirect judgments of the entrepreneurs’ charismatic performances by 98 listeners. The correlations we found replicate previous findings in that a larger F0 range and a higher/lower F0 level are beneficial for a male/female speaker’s perceived charisma. Moreover, LTAS settings that are indicative of a fuller and less breathy voice also led to higher speaker charisma ratings. The same applies to LTF settings that are indicative of a larger body or vocaltract size. The findings are discussed with respect to their implications for measuring and training charismatic speech, traditional rhetoric statements and the definition of charisma.
Original languageEnglish
Title of host publicationProceedings of the 9th International Conference on Speech Prosody 2018
PublisherInternational Speech Communication Association
Publication date13. Jun 2018
Pages359-363
DOIs
Publication statusPublished - 13. Jun 2018
Event9th International Conference on Speech Prosody 2018 - Poznan, Poland
Duration: 13. Jun 201816. Jun 2018

Conference

Conference9th International Conference on Speech Prosody 2018
Country/TerritoryPoland
CityPoznan
Period13/06/201816/06/2018

Keywords

  • Charisma
  • LTF
  • LTAS
  • entrepreneur
  • voice
  • F0
  • Entrepreneur
  • Voice

Fingerprint

Dive into the research topics of 'The acoustic fingerprint of a charismatic voice - Initial evidence from correlations between long-term spectral features and listener ratings'. Together they form a unique fingerprint.

Cite this