On the Use of a Serious Game for Recording a Speech Corpus of People with Intellectual Disabilities

15976

Authors

nan

Abstract

AbstractThis paper describes the recording of a speech corpus focused on prosody of people with intellectual disabilities. To do this, a video game is used with the aim of improving the user’s motivation. Moreover, the player’s profiles and the sentences recorded during the game sessions are described. With the purpose of identifying the main prosodic troubles of people with intellectual disabilities, some prosodic features are extracted from recordings, like fundamental frequency, energy and pauses. After that, a comparison is made between the recordings of people with intellectual disabilities and people without intellectual disabilities. This comparison shows that pauses are the best discriminative feature between these groups. To check this, a study has been done using machine learning techniques, with a classification rate superior to 80%.

Keywords

nan

Full Content

On the Use of a Serious Game for Recording a Speech Corpus of People with Intellectual Disabilities Mario Corrales-Astorgano1, David Escudero-Mancebo1, Yurena Guti´errez-Gonz´alez2, Valle Flores-Lucas3, C´esar Gonz´alez-Ferreras1, Valent´ın Carde˜noso-Payo1 1Department of Computer Science, University of Valladolid, Spain 2Department of Spanish Philology, Universitat Aut`onoma de Barcelona, Spain 3Department of Psychology, University of Valladolid, Spain macoas.5@gmail.com,{descuder,cesargf,valen}@infor.uva.es, yurenagg@gmail.com,vflores@psi.uva.es Abstract This paper describes the recording of a speech corpus focused on prosody of people with intellectual disabilities. To do this, a video game is used with the aim of improving the user’s motivation. Moreover, the player’s profiles and the sentences recorded during the game sessions are described. With the purpose of identifying the main prosodic troubles of people with intellectual disabilities, some prosodic features are extracted from recordings, like fundamental frequency, energy and pauses. After that, a comparison is made between the recordings of people with intellectual disabilities and people without intellectual disabilities. This comparison shows that pauses are the best discriminative feature between these groups. To check this, a study has been done using machine learning techniques, with a classification rate superior to 80%. Keywords: Pronuntiation Training, Intellectual disabilities, Down Syndrome, Prosody 1. Introduction People with intellectual disability (ID), including Down syndrome people (DS), show a wide range of language difficulties (Cleland et al., 2010). The speech of ID peo- ple usually presents multiple disorders affecting the differ- ent components of language (syntax, semantics, phonology and pragmatics) (Martin et al., 2009) (Chapman, 1997). In particular, prosody is also affected (Stojanovik, 2011) conditioning their communicative potential. There is some research on linguistic deficits on ID people (Tsakiridou, 2006) but speech corpora that permit to compare and an- alyze the problems of this population are scarce. The CHILDES database (MacWhinney, 2000) or the Alborada corpus (Saz et al., 2010) have recordings of people with different typologies of intellectual disability, but, these cor- pora don’t focus on prosody. It is quite difficult to obtain a representative ID’s speech corpus focused on prosody, among other reasons, because ID people have a variability in their speech competences: some prosodic aspects could be relegated by the difficulties with other linguistic compo- nents like articulatory or syntactic. Other cognitive diffi- culties like their attentional problems or their impairments with the auditory memory make it difficult to obtain lin- guistic materials that prompt their prosodic competences avoiding their other linguistic or cognitive difficulties. In this paper we present a procedure to collect samples of ID’s utterances related with their prosodic competences that are recorded in a controlled scenario. The present work is framed in a research project 1 whose goal is to develop a serious video game for people with intellectual disabilities for they to practice oral pronunci- 1Proyecto Recercaixa 2013ACUP00145 ”¡Juguemos a comu- nicar mejor! La mejora de la competencia pros´odica como v´ıa de integraci´on educativa e inclusi´on social del alumnado con necesidades educativas especiales derivadas de la discapacidad” http://prado.uab.es/recercaixa/es/index.html ation. The main objective of the video game is that users to improve their prosodic skills both in comprehension and …

Content truncated to 500 words

Labeled Text

AbstractThis paper describes the recording of a speech corpus focused on prosody of people with intellectual disabilities. To do this, a video game is used with the aim of improving the user’s motivation. Moreover, the player’s profiles and the sentences recorded during the game sessions are described. With the purpose of identifying the main prosodic troubles of people with intellectual disabilities, some prosodic features are extracted from recordings, like fundamental frequency, energy and pauses. After that, a comparison is made between the recordings of people with intellectual disabilities and people without intellectual disabilities. This comparison shows that pauses are the best discriminative feature between these groups. To check this, a study has been done using machine learning techniques, with a classification rate superior to 80%.

Publication Details

Paper ID:
15976
Year:
2016
Publication:
TIPSTER
Publication Type:
conf
DOI:
nan