Speech recognition is a vital part of a lot of technological advancements in modern times. We now use things like Amazon Echo, Google Maps’s hands-free mode, or even Siri, the virtual assistant on iOS devices, all of which feature speech recognition. However, this technology leaves out people with speech impairments, or people with a strong accent. This is a big issue today, as there are millions of people worldwide who cannot use speech recognition technology because of one reason or another. However, Tel Aviv-based startup Voiceitt is developing an app that would help speech-impaired people to make use of the technology.
Here’s how the Voiceitt app works: the user launches the app, then it asks them to read out short, useful sentences like I’m thirsty, Turn off the lights, etc. The software records the person’s pronunciation and studies it. If the person is unable to speak at all, then they can indicate. After a brief training period, the Voiceitt app can turn the user’s statement into normalized speech, which it outputs in form of audio or text messages. Voiceitt can also be used by people with speech impediments to communicate face-to-face with other people.
An early version of the app will be released next year. Currently, the app is still in beta tests. The company’s pilot customers are hospitals, schools and people who have speech impediments due to a health condition, like people suffering from cerebral palsy, Down syndrome, Parkinson’s disease, or people recovering from a traumatic brain injury or stroke.
Voiceitt CEO Danny Weisberg had this to say about the app:
“I realized how we take for granted the way we communicate by speaking. Losing this is really terrible, one of the hardest aspects of stroke recovery. . . This could be an accessibility extension to speech recognition for anyone, [including] Google, Amazon, Apple, IBM or Microsoft. We’d love to function like a major OEM and work with all the major platforms.”