Search

Technological advancement in the area of synthetic speech has made it increasingly difficult to distinguish quality of speech based solely on intelligibility scores obtained in benign laboratory conditions. Intelligibility scores obtained for natural speech and a high-quality text-to-speech system (DECtalk) are not substantially different. This study examined the perceived intelligibility and speech interference thresholds of DECtalk male and female voices and compared them with data obtained for natural speech. Results revealed that decreasing signal-to-noise levels had more deleterious effects on the perception of DECtalk male and female voices than on the perception of natural speech. Analysis of pattern of phoneme errors revealed that similar general patterns of errors tended to occur in DECtalk and in natural speech. The speech interference test did not demonstrate any significant difference between the DECtalk male and female voices. These results were supported by the absence of a significant difference between DECtalk male and female voices during intelligibility testing at different signal-to-noise ratios.

This study evaluated normally developing preschoolers' imitation of intonation contours modeled in a sentence elicitation task. Three intonation contours (declarative, interrogative, and monotone) were presented to 3- and 5-year-old children. Acoustic analyses using the Visi-Pitch and Apple IIe computer system, along with perceptual ratings, measured the extent to which preschoolers imitated a modeled intonation contour. The results indicated that, as a group, the children were perceived to imitate or partially imitate the preceding intonation contour modeled by the adult. The 5-year-old children imitated modeled contours more frequently than did the 3-year-old children. In contrast, the 3-year-old children primarily partially imitated the intonation contours. Further analysis revealed that between-group differences were due largely to accuracy in imitating the interrogative contour. These findings establish reference data on normal children and implicate the use of both acoustic and perceptual analyses in order to develop and interpret intervention strategies aimed at children with prosodic problems.

Many English-speaking children with specific language impairment have been found to be especially weak in their use of grammatical morphology. In a separate literature, many children meeting the same subject description have shown significant limitations on tasks involving the perception of rapid acoustic changes. In this study, we attempted to determine whether there were parallels between the grammatical morphological limitations of children with specific language impairment and their performance profiles across several perceptual contrasts. Because most English grammatical morphemes have shorter durations relative to adjacent morphemes in the speech stream, we hypothesized that children with specific language impairment would be especially weak in discriminating speech stimuli whose contrastive portions had shorter durations than the noncontrastive portions. Results from a group of eight children with specific language impairment with documented morphological difficulties confirmed these predictions. Several possible accounts of the observed morphology-perception parallels are offered.

This study examined the perceived intelligibility of synthetic speech. Participants were adults aged 49-69, one group with normal hearing and one group with acquired sensorineural hearing impairment. Word lists were presented in two speech types: DECtalk (a high-quality speech synthesizer) and a natural male speaker. Results revealed differences between groups, with normal-hearing listeners scoring higher than hearing-impaired listeners, and between speech types, with higher scores for natural speech than for synthesized speech. There was no significant interaction of hearing level and speech type.

This study attempted to assess the abilities of 10 normal speakers, five stutterers in therapy, and five stutterers no longer in therapy, to control the time program of repeated utterances. The speech sample comprised repeated sentences, paragraphs, and nursery rhymes, and a finger-tapping task was included as a control. Temporal accuracy was measured. Results suggest that (1) there is a wide range of timing abilities, even among the normal speakers, with considerable overlap between the different groups of speakers; (2) on most of the experimental tasks, normal speakers are more accurate timers than are stutterers; (3) stutterers released from therapy are more accurate timers than are stutterers still in therapy, whenever these groups differ; and (4) subjects' speech timing scores correlate moderately with their tapping scores. These results are discussed in terms of (1) theoretical timing control processes, such as a neural clock for controlling speech segment durations, and a speech motor output buffer, whose capacity may be limited in stutterers, and (2) known effects of rhythmic constraints and respiratory irregularity on fluency.

Young children sometimes make use of unusual phonological patterns even when they already possess the appropriate sound or a suitable substitute in their phonological systems. In this investigation, we attempted to determine whether in such instances unusual sound changes enable children to avoid potential homonymy with other words in their lexicons. Novel words were presented to children, half serving as potential homonyms, half as unlikely homonyms. The children's acquisition of these words was monitored. For a group of normally developing children, unusual sound changes were found to be more frequent in the words with the potential for homonymy. In contrast, a group of children with specific language impairment showed the same degree of unusual usage for both types of words. The findings suggest that children with specific language impairment are especially limited in their ability to capitalize on the phonetic regularities of the language.