The quality of the voice in Tracheoesophageal (TE) speech is determined by the characteristics of the neo-glottis. Individual differences in the functioning of the neo-glottis after treatment cause great variation in the intelligibility and quality of speech. The voice pathology of TE speech is graded into four levels.
Type I - Stable & Harmonic (press 1)
Type II - Stable & At least one harmonic (press 2)
Type III - Unstable or Partly harmonic (press 3)
Type IV - Barely harmonic (press 4)
(press 0 to reset)
Table of the relation between the four types of acoustic signal typing and the perceptual judgment of overall voice quality for 39 speakers (converted to percentages).
| Perceptual judgment of overall voice quality |
| Good Reasonable Poor |
| Type I 70% 40% 0% |
| Type II 45% 45% 10% |
| Type III 20% 35% 45% |
| Type IV 0% 25% 75% |
In Acoustic Signal Typing, the voice characteristics are determined using acoustic analysis of speech. The typing is based on both visual inspection of plots of these analysis parameters and quantitive measures of a short (e.g., 0.1 second) stretch of "stable" speech.
Visual determination of pathology uses displays of:
A quantitative evaluation is based on the analysis of:
These measures are determined on a short segment (around 0.1 second) of speech from the most stable part of a sustained /a/ sound. Pathological categories are defined using (Van As, 2001). See Analysis calculations and displays for details on the commands used.
© R.J.J.H. van Son, December 1, 2011