Acoustic analysis of durational characteristics in normal and articulatory defective adult speakers

The purpose of this study was to investigate the differences between durational characteristics of phonemically accurate productions of vowel + stop consonant + vowel (V₁CV₂) sequences in adult speakers with articulatory defects and adult speakers with normal articulation. Durational measurements on the following subsegments were obtained: steady state of the preconsonantal vowel; transition of the preconsonantal vowel; consonant; transition of the postconsonantal vowel; steady state of the postconsonantal vowel; and total disyllable.

Ten adults with articulation errors sufficient to place them one standard deviation below the mean score for an eight year old child on the Templin-Darley Tests of Articulation comprised the experimental group. Ten adults with no articulation errors and no history of articulation errors comprised the control group. All subjects met the following criteria: : no significant deviation in the structure or function of the oral mechanism; normal hearing; normal voice and fluency; normal intelligence; and negative history of neurologic damage or symptoms. The stimuli used in this investigation were eighteen vowel + stop consonant + vowel sequences embedded in the phrase, "Say the word V₁CV₂ please." The subjects read each of the stimuli five times. These responses were tape recorded and later analyzed on the Voice Identification Unit. All durational measurements made from the broadband spectrograms were obtained at the second formant. The duration measurements were used to infer timing of articulatory movements during the subsegments of the V₁CV₂ sequence.

The major findings of this study were:

1. The duration of the steady state of the preconsonantal vowel (SSV₁) was significantly longer in the articulatory defective speakers than in the normal speakers.

2. The duration of the transition of the preconsonantal vowel (TV₁) was not significantly different in the group of normal speakers and the group of articulatory defective speakers.

3. The duration of the intervocalic consonant (CD) was significantly longer in the group of articulatory defective speakers than in the group of normal speakers.

4. The duration of the transition of the postconsonantal vowel (TV₂

5. The duration of the steady state of the postconsonantal vowel (SSV2) was significantly longer in the group of articulatory defective speakers than in the group of normal speakers.

6. The total disyllable duration (TDD) was significantly longer in the group of articulatory defective speakers than in the group of normal speakers.

7. Two speakers in the articulatory defective group exhibited durational values similar to those exhibited by speakers in the group of normal speakers. One speaker in the group of normal speakers exhibited durational values similar to those obtained by the articulatory defective speakers.

In general, the group of articulatory defective speakers had significantly longer mean durations of all subsegments measured in the ViCV2 sequence with the exception of transition of the preconsonantal vowel (TV) for which there was no significant difference between groups of speakers. These results indicate an involvement in the planning and/or integration of articulatory movements in the group of articulatory defective speakers.

Further research is needed to: (1) investigate the consistency of these results in other phonemic sequences; (2) investigate the consistency of these results in groups of normal speaking and articulatory defective children; and (3) investigate the consistency of these results under imposed rate conditions.