Adobe is known to the general public for Photoshop, an image processing program. But he has a new program up his sleeve that can shake the world. At the AdobeMax conference, he presented a prototype of the Project VoCo program, which they say is Photoshop for sound. Namely, it imitates the color of the voice of the author of the audio recording and can create speech that can be heard as if it were the actual words of the author of the audio recording and not artificially added or changed.
When you hear what he has prepared Photoshop, you won't be able to believe your ears! Literally, because Adobe presented a prototype of the program Project VoCo, which synthesizes speech and voice color. Based on algorithms, it is capable digitally recreate spoken words the author of the sound recording, so much so that we cannot distinguish the original from the artificially altered.
What Photoshop does with photography, VoCo does with sound. VoCo learns the user's speech, and it can be used to record any text that the program reads, as you would pronounce it yourself.
READ MORE: The power of Photoshop - celebrities and celebrities before and after digital photo manipulation
Project VoCo allows you to manipulate an existing audio recording. You can hijack content and add words that were never actually recorded. But no one will notice. In practice, this means that if the player mispronounces or forgets a word, it can be corrected afterwards artificially corrected or added. It will be enough typed word in the program and based on the existing voice, the VoCo will create itself identical sound voice. Although, from a technical point of view, not much is known about the program itself, its potential can be understood from the demonstration.
Adobe accompanied Project VoCo with these words: "When recording voices, dialogue, etc., it often happens that we want only changed a word or two and thus corrected the error or changed the message, the tone..." In addition to the advantages brought by such technology, the first concerns that are far less innocent than those of Photoshop. If possible to that extent synthesize speech, that it cannot be separated from the original, that's it an ideal weapon of manipulation, if it falls into the wrong hands.
Who will we be able to trust then? They will sound recordings in this case even credible? Pandora's box is open and there is no turning back...
More information:
adobe.com
