We use cookies to ensure that we give you the best experience on our website Learn more

Mar 11, 2019

13th World Congress on Brain Injury

Application of Automated Vowel Space Area Analysis to Predict Concussion Status

;

Concussion

Vowel Space Area

Automated Voice Analysis

Speech Analysis

Abstract

Abstract

thumbnail

Keywords

Concussion

Vowel Space Area

Automated Voice Analysis

Speech Analysis

Abstract

New methods for identifying concussions are being researched in an attempt to find inexpensive, more reliable protocols. Importantly, there is a general lack of understanding regarding the effect of concussion on speech production and the voice of brain injured individuals especially those who have sustained a concussion [1]. Additionally, there is even less research available comparing the speech and voice characteristics of concussed athletes to that of normal healthy speakers. Few researchers have attempted to analyze the voices of concussed athletes [2]. Those that have to date examined the voices of these athletes suggest that changes in the voice following head injury may be used to identify individuals with concussion. More robust voice analysis, less affected by recording noise and speech variability is needed. The current research proposes and outlines the pilot research involving formant analysis of athletes following concussion. Are there measurable differences in the formant frequencies of normally healthy athletes versus athletes who have a concussion. Participants consisted of 92 athletes from throughout the Midwest. Several formant analyses were performed on vowels automatically extracted from connected speech of concussed athletes before and after sustaining a concussion. Formant frequencies were then extracted from these vowels and vowel space area calculations were performed [3]. F1-F2 trajectories can potentially predict articulatory dynamic specific to pre- and post-concussion. F2 predicted concussion status: paired t-test (t(83) = -2.0580, p = 0.04 (<0.05)). F1-F2 trajectories suggest lack of control over tongue movements during vowel utterance after concussion. Types of discriminative features vary among individuals. Separation of concussed voice from non-concussed one requires investigating a multidimensional acoustic space

Discover over 20,000 new abstracts, posters and presentations from leading academic conferences every month. Stay on top of the latest findings, methodologies and discussions happening in your research field around the world.

Company

Legal

Follow us

© Copyright 2019 Morressier GmbH. All rights reserved.

© Copyright 2019 Morressier GmbH.
All rights reserved.