Acoustic and perceptual analysis of speech adaptation to an artificial palate

Use of electropalatography in the treatment of speech disorders in children with Down syndrome: a randomized controlled trial

BACKGROUND

Electropalatography (EPG) records details of the location and timing of tongue contacts with the hard palate during speech. It has been effective in treating articulation disorders that have failed to respond to conventional therapy approaches but, until now, its use with children and adolescents with intellectual/learning disabilities and speech disorders has been limited.

AIMS

To evaluate the usefulness of EPG in the treatment of speech production difficulties in children and adolescents with Down syndrome (DS) aged 8-18 years.

METHODS & PROCEDURES

A total of 27 children with DS were assessed on a range of cognitive and speech and language measures and underwent additional EPG assessment. Participants were randomly allocated to one of three age-matched groups receiving either EPG therapy, EPG-informed conventional therapy or 'treatment as usual' over a 12-week period. The speech of all children was assessed before therapy using the Diagnostic Evaluation of Articulation and Phonology (DEAP) and reassessed immediately post- and 3 and 6 months post-intervention to measure percentage consonants correct (PCC). EPG recordings were made of the DEAP assessment items at all time points. Per cent intelligibility was also calculated using the Children's Speech Intelligibility Measure (CSIM).

OUTCOMES & RESULTS

Gains in accuracy of production immediately post-therapy, as measured by PCC, were seen for all groups. Reassessment at 3 and 6 months post-therapy revealed that those who had received therapy based directly on EPG visual feedback were more likely to maintain and improve on these gains compared with the other groups. Statistical testing showed significant differences between groups in DEAP scores across time points, although the majority did not survive post-hoc evaluation. Intelligibility across time points, as measured by CSIM, was also highly variable within and between the three groups, but despite significant correlations between DEAP and CSIM at all time points, no statistically significant group differences emerged.

CONCLUSIONS & IMPLICATIONS

EPG was an effective intervention tool for improving speech production in many participants. This may be because it capitalizes on the relative strength of visual over auditory processing in this client group. The findings would seem to warrant an increased focus on addressing speech production difficulties in current therapy.

Comparing the effects of Essix and Hawley retainers on the acoustics of speech

Objective

The aim of this study was to compare the effects of two retainer types (Essix and Hawley) on speech performance.

Subjects and methods

The speech articulation of 30 patients was evaluated prospectively. Five patients did not appear during the follow-up periods. The patients were randomly divided into retention groups by treatment allocation cards as Essix and Hawley. The Essix group included 13 participants with a mean age of 15.3±2.4 years; the Hawley group included 12 participants with a mean age of 16.3±2.56 years. Speech sound assessments were performed on the first day and 1 week, 4 weeks, and 3 months later. On the first day, the assessments were conducted prior to inserting the retainers, immediately after maxillary and mandibular retainer application, individually, and with both retainers applied. The acoustic analyses were obtained using spectral and temporal parameters.

Results

Statistical analyses were performed with IBM SPSS for Windows, version 20. A P value less than 0.05 was considered statistically significant. The most apparent changes were found in the [a] vowel in the Hawley group, the [e] vowel in the Essix group, and the [u] vowel in both groups (P < 0.05). While the number of affected consonant-vowel couples in the Essix group was low, alterations were common in the Hawley group. There was a statistically significant difference (P < 0.05) in voice onset time of the [d] sound between the groups.

Limitations

The trial had a small sample size and a short follow-up period.

Conclusions

The Hawley retainer affected articulatory movements in consonant-vowel combinations more prominently than the Essix retainer did. Voice onset time of the consonant [d] in the Hawley group was shorter than normal, indicating rapid articulatory movement in the alveolar region.

Speech Adaptation to Kinematic Recording Sensors: Perceptual and Acoustic Findings

PURPOSE

This study used perceptual and acoustic measures to examine the time course of speech adaptation after the attachment of electromagnetic sensor coils to the tongue, lips, and jaw.

METHOD

Twenty native English speakers read aloud stimulus sentences before the attachment of the sensors, immediately after attachment, and again 5, 10, 15, and 20 min later. They read aloud continuously between recordings to encourage adaptation. Sentence recordings were perceptually evaluated by 20 native English listeners, who rated 150 stimuli (which included 31 samples that were repeated to assess rater reliability) using a visual analog scale with the end points labeled as "precise" and "imprecise." Acoustic analysis began by segmenting and measuring the duration of the fricatives /s/ and /ʃ/ as well as the whole sentence. The spectral center of gravity and spectral standard deviation of the 2 fricatives were measured using Praat. These phonetic targets were selected because the standard placement of sensor coils on the lingual surface was anticipated to interfere with normal fricative production, causing them to become distorted.

RESULTS

Perceptual ratings revealed a decrease in speech precision after sensor attachment and evidence of adaptation over time; there was little perceptual change beyond the 10-min recording. The spectral center of gravity for /s/ decreased, and the spectral standard deviation for /ʃ/ increased after sensor attachment, but the acoustic measures showed no evidence of adaptation over time.

CONCLUSION

The findings suggest that 10 min may be sufficient time to allow speakers to adapt before experimental data collection with Northern Digital Instruments Wave electromagnetic sensors.

Motor equivalent strategies in the production of German /∫/ under perturbation

The German sibilant /∫/ is produced with a constriction in the postalveolar region and often with protruded lips. By covarying horizontal lip and tongue position speakers can keep a similar acoustic output even if the articulation varies. This study investigates whether during two weeks of adaptation to an artificial palate speakers covary these two articulatory parameters, whether tactile landmarks have an influence on the covariation and to what extent speakers can foresee the acoustic result of the covariation without auditory feedback. Six German speakers were recorded with EMA. Four of them showed a covariation of lip and tongue, which is consistent with the motor equivalence hypothesis. The acoustic output, however, does not stay entirely constant but varies with the tongue position. The role of tactile landmarks is negligible. To a certain extent, speakers are able to adapt even without auditory feedback.

Electropalatographic specification of Croatian fricatives /s/ and /z/

Electropalatographic specification of alveolar fricatives in Croatian is aimed at providing speech therapists with normative data about the range of acceptable productions of /s/ and /z/ in adult speakers of Croatian. Four variables were analysed: place of articulation, total contact, groove width and hold phase duration. Intra- and inter-speaker variability for each variable was analysed. Lingual palatal cues for voicing difference were also quantified and discussed. Results show that Croatian /s/ and /z/ are alveolar and not dental as previously reported. The comparison between the voiced and the voiceless fricative shows that durational measures provide the best differentiation. The voiceless counterpart is significantly longer. The difference between voiced and voiceless is also found in the total contact, with /z/ having more contact in the anterior four rows of electrodes, while /s/ has more contact in the posterior four rows of electrodes. This difference is also reflected in the anterior and the posterior groove widths. Possibilities of using these results as normative data for the diagnosis and treatment of atypical articulation of /s/ and /z/ are discussed.

An EPG analysis of /t/ in young people with Down's syndrome

Many studies have pointed to impaired speech intelligibility in young people with Down's syndrome (DS). Some have attributed these problems to delayed phonological development, while others have identified disordered speech patterns, which could be related to a dyspraxic element in their speech. This study uses electropalatography (EPG) to examine the speech of 25 young people with DS, focusing on their production of the obstruent /t/. For the EPG analysis, participants produced the target obstruent in the word 'toe', repeated 10 times. An investigative analysis was carried out. A new descriptive taxonomy of EPG error patterns was developed and these errors were related to perceptually based transcriptions. The measures are discussed in relation to current knowledge of the anatomical and physiological characteristics of DS.

A perceptual and electropalatographic study of /integral/ in young people with Down's syndrome

Speech production in young people with Down's syndrome has been found to be variable and inconsistent. Errors tend to be more in the production of sounds that typically develop later, for example, fricatives and affricates, rather than stops and nasals. It has been suggested that inconsistency in production is a result of a motor speech deficit. Late acquired fricatives such as /s/ and /integral/ are complex articulations, which may require more precise motor programming and may therefore show highly inconsistent productions. Other factors potentially affecting speech production in this population are abnormal palatal structure, hearing loss, and hypotonia. A group of 20 young people with Down's syndrome were recorded using Electropalatography (EPG), reading a wordlist containing the phrase 'a sheep'. The wordlist contained seven other phrases and was repeated 10 times. Eight typically developing, cognitively matched children and eight adults were also recorded producing the same data set. Articulatory (EPG pattern analysis) and perceptual analyses of the 10 productions of /integral/ were carried out. /integral/ production was found to be inconsistent in the young people with Down's syndrome, with more errors both in the auditory analysis and articulatory analysis than in the typical sample, which may be due to a motor programming or motor control problem. There were a greater number of errors in the EPG analysis than in the perceptual analysis. This suggests that some young people with DS were able to produce perceptually acceptable /integral/ with atypical EPG patterns. The use of typical, adult-modelled /integral/ EPG patterns in therapy may be inappropriate for some children with DS who present with atypical palatal structures.