Adobe has been able to render unavoidable with software such as Photoshop, a tool used in the world for retouching images. So when the publisher announced a new project, everyone is listening. During the conference, Adobe Max that took place a few days ago, Adobe has lifted the veil on its Project Voco. A sort of Photoshop for the voice, which would rework an audio file by changing the words.
Adobe is working on a revolutionary software for the voice
Adobe, the software specialist publishing of any kind, working on a new project. This project, called the Voco, is in development with the collaboration of the university of Princeton. Voco is a audio editing software like no other, since it would modify a sentence at will.
This program would indeed be able to parse a sentence said by a person to soak up his voice and its intonation. He would then add words which would then be pronounced as if they had been registered by the person himself. To do this, the software must, however, rely on a recording of the voice of this person, minimum 20 minutes. “We have developed a technology that allows to simply type the words you want to change in the voice synthesis “, says a spokesperson for Adobe. Adding : “the algorithm does the rest and generates words that seem to come from the original person. “
The Project Voco in action – screen capture of a video conference Adobe Max, published by the editor on YouTube
A demonstration bluffing
During the presentation of this project, Adobe has made a demonstration of his Project Voco. The effect is stunning, it is impossible to differentiate between the true voice of the voice synthesis. After the possibility of fake videos and images, it will be possible with such a tool rig of speech. We could soon no longer be able to entrust in the audio recordings.
No comments:
Post a Comment