Search

Purpose: This study examined the effects of Speech Intelligibility Treatment (SIT) on intelligibility and naturalness of narrative speech produced by francophone children with dysarthria due to cerebral palsy. Method: Ten francophone children with dysarthria were randomized to one of two treatments, SIT or Hand-Arm Bimanual Intensive Therapy Including Lower Extremities, a physical therapy (PT) treatment. Both treatments were conducted in a camp setting and were comparable in dosage. The children were recorded pre- and posttreatment producing a story narrative. Intelligibility was measured by means of 60 blinded listeners' orthographic transcription accuracy (percentage of words transcribed correctly). The listeners also rated the children's naturalness on a visual analogue scale. Results: A significant pre- to posttreatment increase in intelligibility was found for the SIT group, but not for the PT group, with great individual variability observed among the children. No significant changes were found for naturalness ratings or sound pressure level in the SIT group or the PT group posttreatment. Articulation rate increased in both treatment groups, although not differentially across treatments. Conclusions: Findings from this first treatment study on intelligibility in francophone children with dysarthria suggest that SIT shows promise for increasing narrative intelligibility in this population. Acoustic contributors to the increased intelligibility remain to be explored further. [ABSTRACT FROM AUTHOR]

Abstract: In human languages, different speech sounds can be contextual variants of a single phoneme, called allophones. Learning which sounds are allophones is an integral part of the acquisition of phonemes. Whether given sounds are separate phonemes or allophones in a listener's language affects speech perception. Listeners tend to be less sensitive to acoustic differences between sounds that are allophones. This study investigated the mechanisms behind the learning of allophones by looking at adults’ sensitivity to acoustic differences between two unfamiliar sounds when they were exposed to input in which the sounds behave like separate phonemes versus allophones in terms of their contextual distribution. The results of two experiments showed that adults became less sensitive to acoustic differences between two unfamiliar sounds after being exposed to input in which the sounds were in complementary distribution and that the emergence of allophonic perception was constrained by the phonetic naturalness of complementary distribution. [ABSTRACT FROM AUTHOR]

Purpose: The purpose of this study was to examine the effects of an internally versus externally cued speech task on perceived understandability and naturalness in speakers with Parkinson’s disease (PD) and cerebellar disease (CD). Method: Sentences extracted from a covertly recorded conversation (internally cued) were compared to the same sentences read aloud (externally cued) by speakers with PD and a clinical comparison group of speakers with CD. Experienced listeners rated the speech samples using a visual analog scale for the perceptual dimensions of understandability and naturalness. Results: Results suggest that experienced listeners rated the speech of participants with PD as significantly more natural and more understandable during the reading condition. Participants with CD were also rated as significantly more understandable during the reading condition, but ratings of naturalness did not differ between conversation and reading. Conclusions: Speech tasks can have a pronounced impact on perceived speech patterns. For individuals with PD, both understandability and naturalness can improve during reading tasks versus conversational tasks. The speech benefits from reading may be attributed to several mechanisms, including possible improvement from an externally cued speech task. These findings have implications for speech task selection in evaluating individuals with dysarthria. [ABSTRACT FROM AUTHOR]

The challenge to enhance the naturalness and efficiency of spoken language man-machine interface, emotional speech identification and its classification has been a predominant research area. The reliability and accuracy of such emotion identification greatly depends on the feature selection and extraction. In this paper, a combined feature selection technique has been proposed which uses the reduced features set artifact of vector quantizer (VQ) in a Radial Basis Function Neural Network (RBFNN) environment for classification. In the initial stage, Linear Prediction Coefficient (LPC) and time-frequency Hurst parameter ( pH) are utilized to extract the relevant feature, both exhibiting complementary information from the emotional speech. Extensive simulations have been carried out using Berlin Database of Emotional Speech (EMO-DB) with various combination of feature set. The experimental results reveal 76 % accuracy for pH and 68 % for LPC using standalone feature set, whereas the combination of feature sets, ( LP VQC and pH VQC) enhance the average accuracy level up to 90.55 %. [ABSTRACT FROM AUTHOR]

Despite the importance of speech naturalness to treatment outcomes, little research has been done on what constitutes speech naturalness and how to best maximize naturalness in relationship to other treatment goals like intelligibility. This study investigated the speech naturalness ratings of individuals with dysarthria and the associated perceptual correlates of highly natural and unnatural speech. Four speakers with hypokinetic dysarthria secondary to Parkinsons disease were recorded and rated for naturalness by 69 students in Communication Disorders. Students were presented with 436 speech samples and asked to provide speech naturalness ratings on a 1-9 Likert scale. After rating speech samples, subjects listed perceptual cues associated with samples rated most and least natural and weighted each cue on a visual analog scale. The data on naturalness ratings showed that spontaneous speech was rated the least natural on average, while sentences from a short story were rated slightly more natural and individually read sentences were rated the most natural of all of the utterance types. Thirteen themes emerged from the perceptual cues collected. Of the 13 themes, intelligibility was rated significantly more important than other cues in highly natural speech and intelligibility and articulation were rated significantly more important than other cues in highly unnatural speech. [ABSTRACT FROM AUTHOR]

Objectives: To assess the perceived pitch and naturalness of popular music by cochlear implant (CI) users.Methods: Eleven experienced post-lingually deafened adult CI users rated the pitch, naturalness, and clarity of a popular song with 10 frequency allocation settings, including the default. The alternative settings all had logarithmic frequency spacing and frequency shifts of less than one octave compared with the default map. For maps which were perceived as having incorrect pitch, participants adjusted the pitch of the song in real time using a slider, in order to normalize it, and the amount of adjustment was recorded.Results: The default map was rated as having close to correct pitch. Naturalness rating was negatively correlated with basal shift from a baseline logarithmic map, which was the same as the default map for basal electrodes (R(2) = 0.77). Ratings of the clarity of the lyrics were adversely affected by basal shift. The majority of participants were able to rate and adjust pitch appropriately. The frequency shift in the map was highly correlated with participants' adjustments of the pitch slider (R(2) = 0.94), but the adjustments were less than expected for the majority of participants.Discussion: The pitch ratings for the default allocation suggest that participants have acclimatized to their processors' frequency allocations. Adjustment of the pitch of the song was possible for the majority and suggested that all but one participant was experiencing frequency compression. Expansion of the frequency allocation might help to alleviate this.Conclusion: Adjustment of the pitch of a popular song could be helpful for tuning CIs. [ABSTRACT FROM AUTHOR]

The memory space and bandwidth may be limited in speech signal communication. To solve the problem, a low rate speech coding and decoding system has been developed to realize speech communication with a high condensation. This system is composed of micro-controller STM32, special speech processor TR600, speech codec chip TLV320AIC10, speech acquisition and playback circuit. This design can code and decode speech in 0. 6 Kbps, 0. 8 Kbps, 1. 2 Kbps and 2. 4 Kbps. The tests show that the system' s reconstructive speech could maintain high naturalness, articulation and intelligibility, hence it reflects the design is feasible, stable and valuable. [ABSTRACT FROM AUTHOR]

Abstract: The quality of current commercial speech synthesis systems is now so high that system improvements are being made at subtle sub- and supra-segmental levels. Human perceptual evaluation of such subtle improvements requires a highly sophisticated level of perceptual attention to specific acoustic characteristics or cues. However, it is not well understood what acoustic cues listeners attend to by default when asked to evaluate synthetic speech. It may, therefore, be potentially quite difficult to design an evaluation method that allows listeners to concentrate on only one dimension of the signal, while ignoring others that are perceptually more important to them. The aim of the current study was to determine which acoustic characteristics of unit-selection synthetic speech are most salient to listeners when evaluating the naturalness of such speech. This study made use of multidimensional scaling techniques to analyse listeners’ pairwise comparisons of synthetic speech sentences. Results indicate that listeners place a great deal of perceptual importance on the presence of artifacts and discontinuities in the speech, somewhat less importance on aspects of segmental quality, and very little importance on stress/intonation appropriateness. These relative differences in importance will impact on listeners’ ability to attend to these different acoustic characteristics of synthetic speech, and should therefore be taken into account when designing appropriate methods of synthetic speech evaluation. [Copyright &y& Elsevier]