Skip to main content

Voiceitt aims to help stroke survivors talk to smart homes

Voiceitt aims to help stroke survivors talk to smart homes


The startup is integrating its nonstandard speech recognition technology into voice-activated devices

Share this story

Waveform by Michael Altemark (Flickr)

Smart home tech could someday be updated with software that makes it easier for people with speech-related medical conditions to communicate with their devices. At CES, the Israeli startup Voiceitt said that it was incorporating its nonstandard speech recognition program into voice-activated devices. The addition could allow people who struggle to communicate to direct and use those systems. 

“All indicators are pointing towards voice technology continuing to grow, so the accessibility component would not just be nice to have, but really important,” Sara Smolley, Voiceitt co-founder, told The Verge.

Voiceitt’s smart home system can’t understand everything people with nonstandard speech say. Users train the system with their own voice and with key phrases like “turn the light on,” which lets it learn each person’s specific vocal patterns.

Dozens of medical conditions, including cerebral palsy, Parkinson’s disease, and stroke, cause people to have trouble communicating: although they know what they want to say, they don’t have as much control over the muscles used to talk, making their speech hard to understand. 

Many people who have these medical conditions also have motor challenges, and can’t move around on their own. They would benefit from voice-activated technologies like smart homes, but standard systems often can’t understand their speech. Improving popular smart devices’ accuracy with nonstandard speech is an important goal to help bring technology to people who may benefit most.

Voiceitt started with a translator application, currently in beta testing, that is designed to help people with nonstandard speech communicate with their friends, family, and caregivers. Users record themselves saying a variety of phrases. Then, the application uses the recordings and information gathered from its database of nonstandard speech samples to learn to interpret how they speak. When they speak into the app, it provides a synthesized audio translation as well as text on-screen. 

Smolley says that Voiceitt has one of the largest existing databases of nonstandard voices. They started building it during an initial crowdsourcing campaign that asked people to donate their voices. The company plans to continue to add voices to the database as people use the app.

While the personal translator application is still in a beta version, the company wanted to expand the applications of their technology, Smolley says. “We learned that in addition to in person communication, we could do a lot more for people: we could help them talk to their machines,” she says.

Voiceitt has received funding from the Alexa Fund, Amazon’s venture capital fund for voice technology innovation, and has participated in accelerator programs with both Google and Amazon. 

Additional reporting contributed by Dieter Bohn.