Apple’s Voice Tool Digitizes Speech

Losing communication ability is one of the many challenges people face when diagnosed with amyotrophic lateral sclerosis (ALS).  Because the condition affects the muscles in the throat and mouth, many patients experience increasing difficulty with speech until they can no longer talk.  Speech synthesis and voice banking – creating a digitized version of one’s voice, have developed into essential tools to help people with ALS communicate.  According to a recent Fast Company report, soon people will be able to more easily bank and use their voice with a new accessibility feature Apple plans to launch later this year. 

Until now, digitizing your voice was a time-consuming and arduous task that many people with an ALS diagnosis postpone when there are many other concerns to navigate.  Personal Voice, however, will only require 15 minutes of spoken phrases for training, which can be broken into smaller recording sessions if necessary.   The voices produced from these sessions work in Apple’s apps and alternative communication apps (AAC) including AssistiveWare

Thanks to advances in artificial intelligence, Personal Voice will allow users to turn their typed text into digitized speech using their own voice.  The initial processing of speech will take some time and may require overnight while the device is charging to complete.  Development of the Personal Voice tool included the collaboration of team members living with ALS, ensuring the product would meet the needs of its users. 

Additional accessibility features from Apple are expected to be released soon, including Live Speech which allows users to generate synthesized speech in apps like FaceTime and for in-person conversations.  Detection Mode allows blind or low-vision users to utilize their phone’s camera to scan text as they navigate daily activities, reading aloud any signs or labels – or identifying a door, object or person.  Assistive Access is another adaptive Apple feature that streamlines popular iPhone and iPad features, making the devices easier for people with cognitive difficulties. 

In creating a platform to give digitized voices greater accuracy, developers worked hard to take accent, tone, inflection and cadence into account when reproducing an individual’s voice.  Apple also took measures to ensure privacy for users, setting up safeguards to prevent a replica of one’s voice from being accessed by someone else.