Skip to main content
Skip to main content

Language Science Lunch Talk - José Ortiz

jose ortiz

Language Science Lunch Talk - José Ortiz

Maryland Language Science Center | Hearing and Speech Sciences Thursday, December 12, 2024 12:15 pm - 1:30 pm HJ Patterson, 2130

Automated Language Sample Analysis for Children with Developmental Language Disorder

Abstract: Automatic speech recognition (ASR) has the potential to improve language sample analysis accuracy and efficiency, though its performance for children with Developmental Language Disorder (DLD) is not well understood. This study uses a dataset of language samples from a corpus of 91 participants with typical language and 17 with DLD, aged 4;0 to 5;11, to examine the concurrent validity and diagnostic accuracy of metrics derived from ASR. The results offer insights into the feasibility of automated language sample analysis by comparing the diagnostic accuracy of both automated and manual transcription methods.

Lunch served at 12:15 PM. 

About: José Ortiz is an Assistant Professor in the Department of Hearing and Speech Sciences at the University of Maryland. He serves as the Director of the Language-Learning Early Advantage Program (LEAP), and the Certificate in Bilingual Speech-Language Pathology. He received a B.A. in Linguistics & Psychology from the University of Connecticut in 2004, an M.A. in Speech-Language Pathology from the University of Massachusetts Amherst in 2007, and a Ph.D. in Special Education from the University of Maryland in 2022. His research focuses on issues regarding the identification of language-related disorders in bilingual children, including disproportionality in special education, non-biased assessment, and technology-enhanced service assessment.

Add to Calendar 12/12/24 12:15:00 12/12/24 13:30:00 America/New_York Language Science Lunch Talk - José Ortiz

Automated Language Sample Analysis for Children with Developmental Language Disorder

Abstract: Automatic speech recognition (ASR) has the potential to improve language sample analysis accuracy and efficiency, though its performance for children with Developmental Language Disorder (DLD) is not well understood. This study uses a dataset of language samples from a corpus of 91 participants with typical language and 17 with DLD, aged 4;0 to 5;11, to examine the concurrent validity and diagnostic accuracy of metrics derived from ASR. The results offer insights into the feasibility of automated language sample analysis by comparing the diagnostic accuracy of both automated and manual transcription methods.

Lunch served at 12:15 PM. 

About: José Ortiz is an Assistant Professor in the Department of Hearing and Speech Sciences at the University of Maryland. He serves as the Director of the Language-Learning Early Advantage Program (LEAP), and the Certificate in Bilingual Speech-Language Pathology. He received a B.A. in Linguistics & Psychology from the University of Connecticut in 2004, an M.A. in Speech-Language Pathology from the University of Massachusetts Amherst in 2007, and a Ph.D. in Special Education from the University of Maryland in 2022. His research focuses on issues regarding the identification of language-related disorders in bilingual children, including disproportionality in special education, non-biased assessment, and technology-enhanced service assessment.

HJ Patterson false