Tongue and lip motion patterns in alaryngeal speech

Kristin J. Teplansky, Alan Wisler, Beiming Cao, Wendy Liang, Chad W. Whited, Ted Mau, Jun Wang

Research output: Contribution to journalConference articlepeer-review

Abstract

A laryngectomy is the surgical removal of the larynx which results in the loss of phonation. The aim of this study was to characterize tongue and lip movements during speech produced by individuals who have had a laryngectomy. EMA (electromagnetic articulography) was used to derive movement data from the tongue and lips of nine speakers (four alaryngeal and five typical). The kinematic metrics included movement duration, range, speed, and cumulative path distance. We also used a support vector machine (SVM) to classify alaryngeal and healthy speech movement patterns. Our preliminary results indicated that alaryngeal articulation is longer in duration than healthy speakers. Alaryngeal speakers also use larger lateral tongue movements and move the tongue back at a slower speed than healthy speakers. The results from the SVM model also indicates that alaryngeal articulatory movement patterns are distinct from healthy speakers. Taken together, these findings suggest that there are differences in articulatory behavior that occur after the removal of the larynx. It may be helpful to consider the distinct articulatory motion patterns of alaryngeal speech in clinical practice and in the development of technologies (e.g., silent speech interfaces) that assist to provide an intelligible form of speech for this patient population.

Original languageEnglish (US)
Pages (from-to)4576-4580
Number of pages5
JournalProceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
Volume2020-October
DOIs
StatePublished - 2020
Event21st Annual Conference of the International Speech Communication Association, INTERSPEECH 2020 - Shanghai, China
Duration: Oct 25 2020Oct 29 2020

Keywords

  • Alaryngeal speech
  • Speech kinematics
  • Support vector machine (SVM)

ASJC Scopus subject areas

  • Language and Linguistics
  • Human-Computer Interaction
  • Signal Processing
  • Software
  • Modeling and Simulation

Fingerprint

Dive into the research topics of 'Tongue and lip motion patterns in alaryngeal speech'. Together they form a unique fingerprint.

Cite this