Detection of Mispronunciations and Disfluencies in Children Reading Aloud

Proença, Jorge; Lopes, Carla; Tjalve, Michael; Stolcke, Andreas; Candeias, Sara; Perdigão, Fernando

Publication

Detection of Mispronunciations and Disfluencies in Children Reading Aloud

2017-08-20Conference paper

dc.contributor.author	Proença, Jorge
dc.contributor.author	Lopes, Carla
dc.contributor.author	Tjalve, Michael
dc.contributor.author	Stolcke, Andreas
dc.contributor.author	Candeias, Sara
dc.contributor.author	Perdigão, Fernando
dc.date.accessioned	2025-10-14T10:57:38Z
dc.date.available	2025-10-14T10:57:38Z
dc.date.issued	2017-08-20
dc.description.abstract	To automatically evaluate the performance of children reading aloud or to follow a child’s reading in reading tutor applications, different types of reading disfluencies and mispronunciations must be accounted for. In this work, we aim to detect most of these disfluencies in sentence and pseudoword reading. Detecting incorrectly pronounced words, and quantifying the quality of word pronunciations, is arguably the hardest task. We approach the challenge as a two-step process. First, a segmentation using task-specific lattices is performed, while detecting repetitions and false starts and providing candidate segments for words. Then, candidates are classified as mispronounced or not, using multiple features derived from likelihood ratios based on phone decoding and forced alignment, as well as additional meta-information about the word. Several classifiers were explored (linear fit, neural networks, support vector machines) and trained after a feature selection stage to avoid overfitting. Improved results are obtained using feature combination compared to using only the log likelihood ratio of the reference word (22% versus 27% miss rate at constant 5% false alarm rate).	eng
dc.description.sponsorship	This work was supported in part by Fundação para a Ciência e Tecnologia under the project UID/EEA/50008/2013 (pluriannual funding in the scope of the LETSREAD project at Instituto de Telecomunicações). The authors acknowledge the support given by Microsoft to this project. Jorge Proença is supported by the SFRH/BD/97204/2013 FCT Grant.
dc.identifier.citation	Proença, Jorge & Lopes, Carla & Tjalve, Michael & Stolcke, Andreas & Candeias, Sara & Perdigão, Fernando. (2017). Detection of Mispronunciations and Disfluencies in Children Reading Aloud. 1437-1441. 10.21437/Interspeech.2017-1522
dc.identifier.doi	10.21437/interspeech.2017-1522
dc.identifier.uri	http://hdl.handle.net/10400.8/14255
dc.language.iso	eng
dc.peerreviewed	yes
dc.publisher	ISCA
dc.relation	Instituto de Telecomunicações
dc.relation	SFRH/BD/97204/2013
dc.relation.ispartof	Interspeech 2017
dc.rights.uri	N/A
dc.subject	Children’s speech
dc.subject	Reading disfluencies
dc.subject	Mispronunciation detection
dc.title	Detection of Mispronunciations and Disfluencies in Children Reading Aloud	eng
dc.type	conference paper
dspace.entity.type	Publication
oaire.awardTitle	Instituto de Telecomunicações
oaire.awardURI	http://hdl.handle.net/10400.8/14168
oaire.citation.conferenceDate	2017-08-20
oaire.citation.conferencePlace	Stockholm, Sweden
oaire.citation.endPage	1441
oaire.citation.startPage	1437
oaire.citation.title	INTERSPEECH 2017
oaire.fundingStream	Financiamento do Plano Estratégico de Unidades de I&D - 2013/2015 - OE
oaire.version	http://purl.org/coar/version/c_970fb48d4fbd8a85
person.familyName	Lopes
person.givenName	Carla
person.identifier.ciencia-id	AF14-3048-F510
person.identifier.orcid	0000-0002-5366-0016
relation.isAuthorOfPublication	4dfbaf0a-8c0b-4eaf-b1ad-0e590a5f3524
relation.isAuthorOfPublication.latestForDiscovery	4dfbaf0a-8c0b-4eaf-b1ad-0e590a5f3524
relation.isProjectOfPublication	090e77d3-8476-4972-9e33-6ad71214fa5c
relation.isProjectOfPublication.latestForDiscovery	090e77d3-8476-4972-9e33-6ad71214fa5c