Babies’ voices used to predict autism

The early speech of autistic babies and toddlers found to differ from children without the condition


Diagnosing autism may be about to get easier and quicker with a discovery that babies and toddler who have the condition make different noises to those who don’t.


It seems that autistic toddlers pronounce words differently to their peers. A new vocal analysis system called Language Environment Analysis (LENA) has been developed to recognise these differences. The system costs £130 and scientists claim it predicts the condition with 86% accuracy.

More than 3.1 million sounds made by 232 children aged between 10 months and 4 years were analysed as part of the study.

Early intervention is key in the treatment of autism and can have a dramatic effect on children’s development. It is hoped that this technology will enable doctors to diagnose and refer children with autism to experts more quickly.

“This technology could help paediatricians screen children for autistic spectrum disorder to determine if a referral to a specialist for a full diagnosis is required and get those children into earlier and more effective treatments,” said Professor Steve Warren, from Kansas University.


It’s estimated one in 58 children in the UK has autism. If you’re concerned about your child’s behavior and interaction, find out more in our guides to autism and Asperger syndrome.

Comments ()

Please read our Chat guidelines.