About This Document (IRD-44308).
This Microsoft Word document created by the La Trobe University Inclusive Resources Development team. This document has been created as a transcript of the supplied audio/video and contains only narrative/spoken content. No audio description has been included.
While every care has been taken to accurately transcribe the original material there may still be errors contained in this conversion.
Project Number.
44308.
Client Name.
Graduate Research School.
Title.
3MT.
Publisher.
La Trobe University.
Date of publication.
2016.
Copyright Notice.
Copyright Regulations 1969.
WARNING.
This material has been copied and communicated to you by or on behalf of La Trobe University pursuant to Part VA of the Copyright Act 1968 (the Act). The material in this communication may be subject to copyright under the Act. Any further copying or communication of this material by you may be the subject of copyright protection under the Act.
Do not remove this notice.
Start Transcript.
Lisa Furlong.
We've all had those conversations with Siri, the ones where she calls Aaron instead of Erin, or where she takes offence at a word you didn't actually say. How frustrating is it? For us, very. But for children with speech sound disorders, a speech recognition system like Siri could prove to be very useful.
More than 40 percent of the children paediatric speech pathologists see have a speech sound disorder. A child with a speech sound disorder has difficulty producing the sounds and words. They may change, add or delete sounds for example they might say 'wabbit' for rabbit or 'nake' for snake. These children require frequent effective and intensive therapy. Without this they're at an increased risk of later social and academic difficulties. But for many families accessing speech therapy is difficult due to a workforce shortage of speech pathologists and barriers around cost and socioeconomic status. For families living in rural and remote areas it is even harder. Thirty percent of the country's children live in a rural or remote area where less than one percent of speech pathologists work. So how could a speech recognition system like Siri help these children?
I am working within a larger team to develop a virtual speech pathologist. This will be a technology-based intervention in the form of a mobile application for use by children with speech sound disorders to improve their speech. It will be performance based, the activities and feedback will be guided by the user's performance. And in time fully autonomous allowing children to practise their speech by themselves. We're going to do this through the use of speech recognition technology, which will automatically analyse the child's speech and immediately tell them which words they've said correctly and which words they haven't. Currently there are no other mobile applications using speech recognition technology, which would make the virtual speech pathologist the first of its kind.
My role within the team is to provide design recommendations from a clinical perspective. As part of this I have interviewed speech pathologists to understand how they manage children with speech sound disorders and to get their perspective on the development of a virtual speech pathologist. I have helped the computer engineering team members build their tailored speech recognition system by collecting speech samples of 450 words from 29 children. A selection of these words has been used to train the speech recognition system. In time the system will be tested to see whether it can reliably determine the accuracy of spoken words.
So yes, Siri can be frustrating, but for children with speech sound disorders, a system like Siri could be their key to speech therapy and a brighter future.
Thank you.
End Transcript.