Biggest Tech Companies Collaborate on Speech Accessibility Technology

Tech heavyweights Amazon, Apple, Google, Meta, and Microsoft are teaming up to make computers and mobile devices more accessible to those with speech disabilities. The goal is to render 'voice-first' user interfaces more accessible to those with speech disabilities. The project is happening at the University of Illinois’s Speech Accessibility Project. The Project is "a new research initiative to make voice recognition technology more useful for people with a range of diverse speech patterns and disabilities.”

Speech impediments must be recognized and accessibility be prioritized technologically speaking. “There are millions of Americans who have speech differences or disabilities. Most of us interact with digital assistants fairly seamlessly, but for folks with less intelligible speech, there can be a barrier to access,” said Clarion Mendes, a clinical professor in speech and hearing science and a speech-language pathologist. “This initiative lessens the digital divide for individuals with disabilities. Increasing access and breaking down barriers means improved quality of life and increased independence. As we embark on this project, the voices and needs of folks in the disability community will be paramount as they share their feedback.”

Using a “private, de-identified dataset” of speech samples collected from paid volunteers, the University of Illinois will take that information and use it to help build stronger, more adept machine learning models so as to better understand more diverse speech patterns. The Project is currently focused on American English, and is designed to accommodate people with Parkinson’s, Down’s syndrome, and other conditions where typical, fluent speech can be compromised. The big goal is to make voice-centric products like digital assistants easier to use and accessible for the millions of Americans who have a speech impairment.

As smart speakers like Apple’s HomePod and Amazon’s Echo line have gained in popularity over the last several years, their utility has felt stifled—and the whole product category as a whole exclusionary—because digital assistants Siri and Alexa are naturally built assuming normal speech. Yet as the Illinois researchers rightfully point out on their FAQ page on the Speech Accessibility Project, the dearth of a diverse dataset means the software cannot learn other ways of speaking. To wit, artificial intelligence and machine learning models are only as good as the information they’re fed by humans; the onus is on software engineers and researchers to gather (and thus use) the most representative data possible.

Contact the Dragon Dictation and Assistive Technology Experts at Northwest Ergonomics

The team at Northwest Ergonomics is here to help with Dragon NaturallySpeaking. Our highly trained assistive technology experts will consult with you to understand your specific needs and provide solutions that really work. We help people of all ages including adults in the workplace who have experienced an injury or other disability.

It is important to our caring specialists that you really get the benefit of these wonderful advancements in technology. For that reason we specialize in training and follow-up to make sure that everything is working as effectively as possible for you.

To contact us, use the form on our contact page, or call 206-707-8830.