Apple filed a patent for ``a function to read iMessage with the sender's voice''
Patently Apple, a news site that checks Apple-related patents and trademarks, reports that a patent application from Apple called 'Personalized Voice for Text Messaging' has been published in the US Patent and Trademark Office database. The filed function was to have messages received by iMessage read aloud by the sender's voice.
20230051062 PERSONALIZED VOICES FOR TEXT MESSAGING
(PDF file)
Apple invents Personalized Voices for Text Messaging - Patently Apple
https://www.patentlyapple.com/2023/02/apple-invents-personalized-voices-for-text-messaging.html
iMessage to voice note: Patent to play messages in your voice
https://9to5mac.com/2023/02/16/imessage-to-voice-note/
For modern people who use electronic devices such as smartphones and PCs, exchanging messages is an essential task, especially in modern society where social media and instant messenger tools have become commonplace. However, after receiving such a message, the task of ``reading the sent text'' has become commonplace, and there are few other ways to check the message.
As an alternative to reading, for example, you can have the voice assistant Siri read out the message you received on the iPhone. However, since Siri reads messages from different senders with the same voice and intonation, there is a problem that it is difficult to confirm who sent the message by voice alone.
The patent filed by Apple this time is a system that prepares a personalized voice for each message. Instead of the sender reading the message to the device and sending a recording, the sender's voice is reproduced on the recipient's device and the message is read aloud.
An example presented in the patent application document states, 'Constructing a speech model on the sender's device based on multiple inputs, and checking if the speech model is also sent when sending a message. On the recipient's device. will check if the sender's voice model is also received.If the sender's voice model is also received, the message will be read out in the sender's voice.'
Since the voice model is sent separately rather than attached to the message, sending the voice model does not delay the receipt of the message. Also, if the recipient does not need to read aloud, the point is that it is possible to choose not to receive the voice model.
According to Patently Apple, this patent application was filed in late October 2022, and the inventor is D. Winalski, director of Siri's Text-to-Speech team.
Related Posts:
