×

Synchronizing virtual actor's performances to a speaker's voice

  • US 9,524,081 B2
  • Filed: 04/15/2015
  • Issued: 12/20/2016
  • Est. Priority Date: 05/16/2012
  • Status: Active Grant
First Claim
Patent Images

1. A method for synchronizing performance of a virtual object to performance of a person, comprising:

  • capturing a first audio signal associated with sounds made by the person during a first time period;

    detecting that a first utterance has been spoken by the person using the captured first audio signal;

    determining a speed at which the person spoke the first utterance;

    capturing a second audio signal associated with sounds made by the person during a second time period subsequent to the first time period;

    detecting that a portion of a second utterance has been spoken by the person using the captured second audio signal; and

    displaying, using a display, a sequence of images of the virtual object at a rate corresponding with the determined speed of the first utterance in response to detecting that the portion of the second utterance has been spoken by the person, the sequence of images including a sequence of mouth shape images displayed such that the virtual object appears to speak the second utterance at the determined speed of the first utterance.

View all claims
  • 2 Assignments
Timeline View
Assignment View
    ×
    ×