There are folding. And there is the new generation of folding that Disney plans to develop in the near future. The traditional folding want just translate or improve the quality of actors or actors in a video dialogues; the new Disney folding system puts these same people say things that originally did not say – even make no sense
Confused.? In the video inserted on this page you can find an example of what the new software producer of Hollywood’s capable of. Front of the camera, a volunteer utters a few words in English: “clean swatches.” The sound of this “speech” is, in a first step, synthesized, and then processed by an application that records the movements of the lips and jaws and searches for words that may require the same movements of the mouth.
After the search, the application goes to the overlap of words spoken with synthesized voice and the images of the person talking. Which means that in the first example described here, the mouth of voluntary pronounced “clean swatches,” but the words can be heard “thence to what saw”; “Laches rush to”; «Die swat saw”; or any other combinations of the 9658 words of the software detected as being given with the same movements of the mouth.
According to the BBC, the system is prepared to avoid grammatical inconsistencies – which does not prevent semantic inconsistencies caused by words that do not always make sense. Disney developed the new application in order to become the most credible voiceovers – but admits he still has much work to do to get all the dialogues of a feature film to be folded using this technology.
No comments:
Post a Comment