We noticed you're blocking ads.

Keep supporting great journalism by turning off your ad blocker.

Questions about why you are seeing this? Contact us

Font Size

- Aa +

Fri 11 May 2018 12:40 AM

Font Size

- Aa +

Emirati dialect turned into database to detect disorders

UAE University's Department of Linguistics creates copyrighted corpus of the Emirati Arabic language

Emirati dialect turned into database to detect disorders

The Emirati dialect has been turned into a database to detect language disorders in children, and to benefit researchers from around the world.

A research project at the UAE University’s Department of Linguistics, entitled Emirati Arabic Language Acquisition Corpus (EMALAC), has successfully resulted in its own copyrighted corpus of the Emirati Arabic language, a statement said.

The corpus includes a database of transcribed texts based on six children speaking with their carers at home. Although much work has been done in other languages, including European and Hebrew, not much has been completed in Arabic due to a lack of resources needed to develop a corpus of children speaking in their natural environment.

"In the UAE, there is a lack of Emirati specialists, speech therapists and language disorder experts," said Dr Dimitrios Ntelitheos, Associate Professor at the Department of Linguistics at the United Arab Emirates University.

"In order to understand if a child has a language problem, you have to compare their speech with what children typically produce, but we don’t have that for Emirati."

He added: "There has been no knowledge in Emirati children’s development of speech over time. However, the research will now allow experts in the field to know what to expect from a certain child at a certain age and detect any potential language delay."

A research assistant from the university compiled the database in 2008 over a period of two years when she interacted with her nieces and nephews for half an hour every week, and recorded the sessions. The six children were aged between two and four when the project began.

A total of 41 half-hour recordings were gathered, with almost all six children – three girls and three boys – speaking around 90,000 sentences, including their utterances and language development, over time. Most of the children participated in the majority of the recordings, while two children took part in only half of the available 41 sessions.

The corpus provides snapshots of the developmental stages that child Emirati Arabic goes through, and forms an essential database on which typical and atypical child language development can be mapped.

The university’s Linguistics Department is currently working towards developing other corpora with children who have other language disorders or language impairment, as well as autistic children. The objective is to compare between typically developing children and atypically developing children of the same age, and develop diagnostics tests that specialists can use to detect language delay in children.

The department has proposed a new BSc program in Communication Sciences and Disorders, currently unavailable across the UAE. The aim of the programme is to compile research from the EMALAC project to include in the curriculum, and develop tests for teaching children and adults in language delay.

Arabian Business: why we're going behind a paywall