Accessibility in iOS 17 features Personal Voice for text-to-speech

Home » Accessibility in iOS 17 features Personal Voice for text-to-speech
Accessibility in iOS 17 features Personal Voice for text-to-speech

Apple’s iOS 17 accessibility preview

Apple has previewed some of the accessibility features it is introducing in iOS 17, including one to help with user interface interactions, and another that allows a user to use their own voice for text-to-speech.

Apple will be formally unveiling iOS 17 and iPadOS 17 during WWDC in June, but on Tuesday, it gave a selective sneak peak at some of the changes. In an accessibility function preview, it outlines updates affecting cognitive accessibility, voice, and visual impairments.

“At Apple, we’ve always believed that the best technology is technology built for everyone,” said Apple CEO Tim Cook. “Today, we’re excited to share incredible new features that build on our long history of making technology accessible, so that everyone has the opportunity to create, communicate, and do what they love.”

The first main feature, Assistive Access, aims to help those with cognitive disabilities, by making it easier to not only see what they need on the display, but to select the items too. This includes the use of large high-contrast buttons and large text labels.

There are further customizations available for users who prefer to communicate visually, such as an emoji-only keyboard in Messages and options to record video messages. The Home Screen is also able to be shown with larger icons, in both visual grid-style layouts or text-based rows.

Live Speech and Personal Voice

Live Speech, which will be available on iPhone, iPad, and Mac, will let users type out statements they want read aloud during phone and FaceTime calls, along with in-person meetings. Commonly-used phrases can also be prepared beforehand for quick deployment in conversations.

For those who are at risk of losing their voice, such as people who have recently been diagnosed with ALS, Personal Voice offers a way to produce a voice for text-to-speech purposes that is closer to the user’s own.

After reading aloud a randomized set of text prompts to create 15 minutes of audio, on-device machine learning then uses the data to create the Personal Voice itself. That Personal Voice can then be used with Live Speech to communicate with others using the user’s own voice, not a generic replacement.

Source link

Leave a Reply

Your email address will not be published.