Comparison of vocal tract area function estimated from audio signal and MRI analysis

Megumi Saito, Shunsuke Ishimitsu, Kazutaka Kasai, Kaori Ishii, Izumi Nishio, Kimiko Yamashita, Satoshi Horihata

Research output: Contribution to journalArticlepeer-review

Abstract

Malocclusion caused by the tongue position makes trouble for communication and health. To cure oral habits that underlie such issues, oral myofunctional therapy plays an important role. Unfortunately, physical examinations using X-rays or devices installed in the patient’s mouth should be involved as the side effects currently. Therefore, the objective tongue position evaluation without burden to the patients have been proposed and studied for a long while. Here, we employ an evaluation technique using vocal tract area function that was estimated from the partial auto-correlation (PARCOR) coefficient to calculate intraoral information while reducing the burden on patients. The research results showed that as the difference between the measured value and the estimated vocal tract area function was significant, it should be reduced with a new method.

Original languageEnglish
Pages (from-to)1201-1205
Number of pages5
JournalICIC Express Letters, Part B: Applications
Volume7
Issue number6
Publication statusPublished - 1 Jun 2016

Keywords

  • MRI
  • PARCOR analysis
  • Speech analysis
  • Vocal tract

Fingerprint

Dive into the research topics of 'Comparison of vocal tract area function estimated from audio signal and MRI analysis'. Together they form a unique fingerprint.

Cite this