Apple researching new Text to speech technology

[dgstorm]

A couple of new Apple patents discovered this week, drew attention towards some very interesting technologies Apple is apparently working on. We’re talking about a text-to-speech engine that generates a customized computer generated voice. It does this by reading emails, texts or other forms of written communication.

The patent is called “Voice assignment for text-to-speech-output” and it describes a technology that works with metadata extracted from the text content found on user’s devices. For example, imagine you are reading an email from a Scottish friend who is 30. The voice generated by the computer will read out loud the email and will mimic a Scottish accent and the tone. The TTS will work by converting each word found within the email into distinct sounds of a given language. The TTS will then manage the rhythmic sounds by dividing them in the manner of phrases, clauses and sentences.

Today, a lot of applications exist that are designed to help those suffering from visual deficits but most TTS engines "generate synthesized speech having voice characteristics of either a male speaker or a female speaker. Regardless of the gender of the speaker, the same voice is used for all text-to-speech conversion regardless of the source of the text being converted."

Apple views things differently. Cupertino wants to offer its users a customized experience. The patent filling goes on to explain:

"Providing a speech output that is associated with a speaker profile allows speaker recognition while providing a more enjoyable and entertaining experience for the listener."

By Radu

Source - Apple looking to add character to text-to-speech voices