Automatic Assessment of Prosodic Quality in Down Syndrome: Analysis of the Impact of Speaker Heterogeneity

Abstract

Prosody is a fundamental speech element responsible for communicative functions such as intonation, accent and phrasing, and prosodic impairments of individuals with intellectual disabilities reduce their communication skills. Yet, technological resources have paid little attention to prosody. This study aims to develop an automatic classifier to predict the prosodic quality of utterances produced by individuals with Down syndrome, and to analyse how inter-individual heterogeneity affects assessment results. A therapist and an expert in prosody judged the prosodic appropriateness of a corpus of Down syndrome' utterances collected through a video game. The judgments of the expert were used to train an automatic classifier that predicts prosodic quality by using a set of fundamental frequency, duration and intensity features. The classifier accuracy was 79.3% and its true positive rate 89.9%. We analyzed how informative each of the features was for the assessment and studied relationships between participants' developmental level and results: interspeaker variability conditioned the relative weight of prosodic features for automatic classification and participants' developmental level was related to the prosodic quality of their productions. Therefore, since speaker variability is an intrinsic feature of individuals with Down syndrome, it should be considered to attain an effective automatic prosodic assessment system.

Evaluating the Impact of an Autonomous Playing Mode in a Learning Game to Train Oral Skills of Users With Down Syndrome

“…For this, we used the offline expert annotations as the training dataset. As we showed in [62], SVM outperforms neural networks and decision trees in this task, reaching an accuracy close to 80% (78.8% recall, 0.291 false positive rate and 0.212 false negative rate).…”

El papel de las habilidades prosódicas en el trastorno del desarrollo del lenguaje. Un estudio de caso

“…Este tipo de estudios de caso único nos permiten obtener información específica de un participante con dificultades y analizar en profundidad su perfil, lo cual podría resultar de gran utilidad para plantear intervenciones eficaces en un caso dado atendiendo al perfil de sus dificultades (Horner y Kratochwill, 2012;Rvachew y Matthews, 2017;Wells y Peppé, 2003). Por otro lado, cabe resaltar que aunque existe una literatura creciente en torno al papel de la prosodia en español y diversos trastornos específicos (e.g., Corrales-Astorgano, et al, 2019;Martínez-Castilla, et al, 2012;Martínez-Castilla y Sotillo, 2014), se aprecia una escasez de estudios que investigan el papel de las habilidades prosódicas en el TDL. Por tanto, se hace necesario realizar más estudios sobre el papel de la prosodia en este trastorno.…”

Editorial for Special Issue “IberSPEECH2018: Speech and Language Technologies for Iberian Languages”

“…The article by Corrales-Astorgano et al [13] aims at developing an automatic classifier to predict the prosodic quality of utterances produced by individuals with Down syndrome. To that effect, it considers a classic set of speech features and a supervised classification approach, but paying special attention to the inter-individual heterogeneity of the speakers according to the results obtained in the Spanish version of the Profiling Elements of Prosody in Speech-Communication (PEPS-C) test.…”