Vocal characteristics of normal speakers and stutterers during choral reading

It is well known that stutterers experience significant decrements in their stuttering when they read or speak in unison with another person. Recently, Wingate suggested that the act of choral reading or speaking prompts the individual who is following the model speaker to emphasize vocalization and its continuity throughout the utterance. This modified vocalization may then be viewed as the immediate cause of the stutterers' reduced disfluency. To evaluate this hypothesis, ten stutterers and ten normal speakers were tested in a control and choral reading condition. In the former, subjects read in their habitual manner. In the latter, subjects read in unison with a recording of a normal adult male. Subjects' oral readings were audio-taped and then submitted to spectrographic analysis. Measures of vowel duration, peak vocal SPL and continuity of phonation were made and then treated statistically. The major findings of this study indicated that across the two conditions, both groups failed to modify their vocal SPL and continuity of phonation. They did, however, alter their vowel durations. The normal speakers increased theirs by a statistically insignificant amount, while the stutterers significantly shortened theirs. The results that pertained to vocal SPL and vowel durations seemed a function of each group's scores for these measures in the control condition as compared to the values for the same measures that were generated by the model speaker with whom subjects read in unison in the experimental condition. These and other findings and interpretations are discussed further relative to Wingate's "modified vocalization" hypothesis.

A comparative study of speech perception in young severely retarded children and normally developing infants

The discrimination of minimally paired speech sounds by seven retarded children with a mean age of 3 years, 2 months and a mean IQ of 38.4 was compared with the discrimination performance of eight normally developing 7-month-old infants. Children and infants were tested using the Visually Reinforced Infant Speech Discrimination (VRISD) paradigm in which they were taught to respond with a head turn to a change in a repeating background auditory stimulus. Responses were reinforced by activation of an animated toy. All children proved to be conditionable and both groups evidenced discrimination of the speech contrasts tested. The data suggest that the retarded children have moree difficulty processing a contrast cued by rapid spectral changes (often associated with consonant discrimination) than they do a contrast cued by steady-state spectral information (often associated with the perception of slowly articulated vowels). The normally developing infants did not find rapid spectral cues mor difficult than steady-state cues These results parallel those of Tallal (1976) who found that dynamic cues were specifically difficult for dysphasic children (with normal nonverbal intelligence), but not for linguistically-normal elementary school children.

Computer analysis of hoarseness

The harmonic components in hoarse voice were separated from the noise components by using a small laboratory computer. The ratio of harmonics to noise or S/N ratio, was calculated and compared with the auditory impression for 58 subjects. The calculated results showed a definite correlation to the auditory impression, and it is suggested that this would be a useful method for the quantitative evaluation of hoarseness.

Acquisition of the voicing contrast language-delayed and normal-speaking children

Voicing is a phonological contrast which emerges early in the speech of children. However, the acoustic correlates of the voicing contrast for stop consonants are fairly complex. In the initial position, voicing is cued primarily by the relative timing of articulatory versus laryngeal gestures. In the final position, the duration of the preceding vowel is associated with the voicing contrast of stop consonants. The purpose of this study was to examine the pattern of acquisition of the voicing contrast in the speech of ten children diagnosed as language-delayed in comparison with the acquisition of the voicing contrast by normal speaking children. The language-delayed and normal-speaking children were matched according to mean length of utterance (MLU) and placed in one of Brown's five developmental stages. Each participant was first given a short test, using natural speech, to determine his or her ability to identify minimal pairs differing in the voicing of stop consonants. Those who passed the test were recorded under standard recording conditions repeating 12 test words. The test words contrasted voiced and voiceless stop consonants in initial and final positions. Spectrograms of the three best productions of each word were used to examine voice-onset time for stops in initial position and preceding vowel duration for stops in final position. Although the language-delayed and normal-speaking children showed equivalent linguistic sophistication (as measured by MLU), the language-delayed children's control of the acoustic-phonetic details of the voicing contrast was less mature than that of the normal-speaking children.

Oral rehabilitation by prosthetic restoration after maxillectomy for malignant tumors

In maxillectomy patients, speech intelligibility between before and after insertion of prostheses has been reviewed by many investigators. Speech intelligibility was not related to the size of the defect, but rather to the defect region. Speech intelligibility was considerably improved by the use of prostheses. A great difference could not be found in escape rate from the nasal cavity between before and after insertion of prostheses, but there was a remarkable change in the spirometer. Postmaxillectomy functions are also easily restored by prosthetic oral rehabilitation, necessary to aid surgical procedure.