Proposal for classifying the severity of speech disorder using a fuzzy model in accordance with the implicational model of feature complexity

Abstract

The objective of this study is to create a new proposal for classifying the severity of speech disorders using a fuzzy model in accordance with a linguistic model that represents the speech acquisition of Brazilian Portuguese. The fuzzy linguistic model was run in the MATLAB software fuzzy toolbox from a set of fuzzy rules, and it encompassed three input variables: path routing, level of complexity and phoneme acquisition. The output was the Speech Disorder Severity Index, and it used the following fuzzy subsets: severe, moderate severe, mild moderate and mild. The proposal was used for 204 children with speech disorders who were monolingual speakers of Brazilian Portuguese. The fuzzy linguistic model provided the Speech Disorder Severity Index for all of the evaluated phonological systems in a fast and practical manner. It was then possible to classify the systems according to the severity of the speech disorder as severe, moderate severe, mild moderate and mild; the speech disorders could also be differentiated according to the severity index.

Normative Features of Phoneme Acquisition in Brazilian Portuguese

“…The production accuracy of all 27 phonemes analyzed in the present study was associated with age, while types of school and gender were only associated with the production accuracy of nine and six phonemes, respectively. In this study, as well as in others, the results showed that some phonemes are mastered earlier, such as /m/, /n/, /ɲ/, /p/, /t/, /k/, while phonemes such as /r/-tap, /ʎ/, and complex onsets are mastered later [6,[32][33][34].…”

“…Consonants that are relatively easier to articulate will likely be acquired before those that are more difficult, especially because the children's articulatory apparatus is physiologically different from that of adults, and their speech motor control abilities have not fully developed [34]. As they mature, children develop the cognitive and articulatory skills involved in the acquisition of more complex phonemes [32,37].…”

Modelo de correlações entre consoantes: implicações para a prática clínica

“…However, these aspects have been highlighted as limitations to its application. Furthermore, the fact that the phonemes have been considered only in simple onset position also limits the use of this model in the choice of target segments for phonological therapy and to determine the severity of phonological disorders, as observed in other studies 4,5 .…”

Impact of communication partner familiarity and speech accuracy on parents’ ratings of their child for the Intelligibility in Context Scale: Dutch

“…The most commonly used measure at the Body Functions level is the segmental measure, percentage consonants correct (PCC), proposed by Shriberg and colleagues (Shriberg, Austin, Lewis, McSweeny, & Wilson, 1997;Shriberg & Kwiatkowski, 1982). But, whole word measures are also used, such as percentage of whole-word correct (PWC), proportion of whole-word proximity (PWP), proportion of whole-word variability (PWV) (Ingram, 2002;Ingram & Ingram, 2001), percentage of words understood (PWU) (Gordon-Brannan & Hodson, 2000), as well as the phonological mean length of utterance (pMLU) and combinations of speech characteristics (Brancalioni, Magnago, & Keske Soares, 2012). Research on reliability, sensitivity and specificity of PCC has shown it to be a reliable measure to assess SSD (Johnson et al, 2004), and PCC is effective in discriminating and identifying children with and without SSD (Barrozo, Pagan-Neves, Pinheiro da Silva, & Wertzner, 2017;Shriberg et al, 1997).…”