Soon your iPhone will be able to replicate your voice after 15 minutes of training

(CNN) — Apple advertisement This Tuesday a series of new accessibility tools for iPhone and iPad, including a feature that promises to replicate a user’s voice for phone calls after just 15 minutes of training.

With a forthcoming tool called Personal Voice, users will be able to read text prompts to record audio and have the technology learn their voice. A related feature called Live Speech will use “synthesized speech” to read aloud user-written text during phone calls, FaceTime conversations, and in-person conversations. People will also be able to save commonly used phrases to use during live conversations.

The feature is one of several aimed at making Apple devices more inclusive for people with cognitive, visual, hearing and mobility disabilities. Apple said that people who may have conditions where they lose their voice over time, such as ALS (amyotrophic lateral sclerosis), could benefit most from the tools.

“Accessibility is part of everything we do at Apple,” said Sarah Herrlinger, Apple’s senior director of Global Accessibility Policy and Initiatives, in a company blog post. “These innovative features were designed with feedback from members of disability communities every step of the way, to support a diverse set of users and help people connect in new ways.”

Apple said the features will roll out later this year.

The risks of technology for deepfakes

While these tools have the potential to fill a genuine need, they also come at a time when advances in artificial intelligence have raised alarm bells about bad actors using convincing fake audio and video, known as “deepfakes,” to defraud or misinform. to the public.

In the blog postApple said the Personal Voice feature uses “on-device machine learning to keep users’ information private and secure.”

Other tech companies have experimented with using AI to replicate a voice. Last year, Amazon said it is working on an update to its Alexa system that would allow the technology to imitate any voice, even that of a deceased relative. (The feature has yet to be released.)

In addition to voice features, Apple announced Assistive Access, which combines some of the most popular iOS apps like FaceTime, Messages, Camera, Photos, Music and Phone into a single Calls app. The interface includes high-contrast buttons, large text labels, an option for an emoji-only keyboard, and the ability to record video messages for people who prefer visual or audio communications.

Apple is also updating its Magnifier app for people with vision problems. It will now include a sensing mode to help people better interact with physical objects. The update would allow someone, for example, to hold an iPhone camera in front of a microwave and move their finger across the keyboard while the app labels and announces the text on the microwave buttons.