iOS 17's "Personal Voice" Feature!
0 up · 0 down · 0 ratings
Description
Hey, what's up guys, I'm MKBHD here and this is maybe the wildest new feature I've ever seen added to the iPhone. Okay, so inside of the accessibility settings of the new iSO 17 under speech, there's something called personal voice. Didn't even really talked about this at the keynote but it uses recordings of your actual voice to stitch together a synthesized voice to sound like you and say anything. There's actually a whole process you have to go through where it gives you like 15 minutes of text prompts that you say out loud into the mic. Then a processes everything on your phone over night and you wake up the next day with a new personalized voice where you can type to speak. All right, it so this is what it sounds like to speak on my voice. I really think 30 frames per second is the correct frame rate. Anyone who says otherwise is clearly faking it. It sounds pretty good. It clearly is a little bit robotic so it kind of sounds like talking to someone on the phone with like a poor connection, you know. But still in 15 minutes to have a voice that can do anything is pretty crazy.
The video introduces Apple's iOS 17 Personal Voice feature, unveiled within the accessibility settings under speech. The host explains that Personal Voice creates a synthesized, lifelike voice using recordings of the user’s own voice. The process begins with a 15 minute prompt session where the user speaks a series of texts, which is then processed on-device overnight to produce a personalized voice. The creator demonstrates the feature by typing and having the new voice speak, noting that the sample voice reaches 30 frames per second and sounds fairly natural, albeit with a slight robotic tone similar to a call with a poor connection. Overall, the video emphasizes the novelty and potential of generating a functional voice from user-provided data, capable of speaking any text after setup. The demonstration underscores a rapid, on-device workflow that ends with a usable synthetic voice ready for everyday communication. The segment concludes by acknowledging the impressive scope of having a voice that can say anything after a single setup, highlighting both technical achievement and the transformative implications for accessibility and personal expression.
Topics · technology · accessibility · ai · consumer_electronics
Questions answered
- What is iOS 17 Personal Voice and how does it work?
- Personal Voice is a feature in iOS 17 accessibility settings that creates a synthesized voice using recordings of the user’s own voice. After a 15 minute prompt session where the user reads text aloud, the data is processed on-device overnight to produce a personalized voice capable of speaking any text.
- How long does it take to set up Personal Voice and when can you use it?
- The setup involves a 15 minute recording session, followed by overnight processing on the device, after which the new voice is available to type and have it speak.
- What are some potential benefits and concerns mentioned in the video and comments?
- Benefits include providing a voice for people who may lose their ability to speak, enabling personalized communication. Concerns include privacy and security risks, such as misuse by scammers or impersonation, and the ethical implications of voice replication.