RT info:eu-repo/semantics/conferenceObject
T1 Pronunciation Assessment and Automated Analysis of Speech in Individuals with Down Syndrome: Phonetic and Fluency Dimensions
A1 Corrales-Astorgano, Mario
A1 González-Ferreras, César
A1 Escudero-Mancebo, David
A1 Aguilar, Lourdes
A1 Flores-Lucas, Valle
A1 Cardeñoso-Payo, Valentín
A1 Vivaracho-Pascual, Carlos
K1 speech disorders, pronunciation assessment, disfluency detection, Down syndrome
AB In this study, we analyze the potential use of an annotated corpus to identify various dimensions of speech quality, including phonetics and fluency, in individuals with Down syndrome, enabling the development of automated assessment systems. Two experiments were conducted: for phonetic evaluation, we used the Goodness of Pronunciation (GoP) metric with an automatic segmentation system and correlated results with a speech therapist’s evaluations, showing a positive trend despite not notably high correlation values. For fluency assessment, deep learning models like wav2vec were used to extract audio features, and an SVM classifier trained on a fluency-focused corpus categorized the samples. The outcomes highlight the complexities of evaluating such phenomena, with variability depending on the specific type of disfluency detected.
PB António Teixeira, Carlos Martinez-Hinarejos, Eduardo Lleida and Dayana Ribas
YR 2024
FD 2024
LK https://uvadoc.uva.es/handle/10324/82147
UL https://uvadoc.uva.es/handle/10324/82147
LA eng
NO António Teixeira, Carlos Martinez-Hinarejos, Eduardo Lleida and Dayana Ribas. IberSPEECH 2024. Aveiro, Portugal. p. 26-30
DS UVaDOC
RD 01-feb-2026