Systems and methods for selective rate of speech and speech preferences for text to speech synthesis

  • US 8,352,268 B2
  • Filed: 09/29/2008
  • Issued: 01/08/2013
  • Est. Priority Date: 09/29/2008
  • Status: Active Grant
  • ×
    • Pin Icon | RPX Insight
    • Pin
First Claim
Patent Images

1. A method for customizing delivery of synthesized speech, the method comprising:

  • generating a speech segment from one or more text strings describing or identifying a media asset having audio data distinct from the generated speech segment;

    obtaining user input requesting a variation in speech delivery accompanying the media asset;

    in response to the user input, customizing the speech segment by modifying selected portions of the speech segment at a server device, wherein the customizing further comprises;

    automatically detecting one or more repeated portions in the speech segment; and

    automatically modifying the speech segment by performing one or more of;

    (1) omitting at least one of the repeated portions from the speech segment, (2) using faster speech patterns for at least one of the repeated portions, (3) shortening breaks between words in at least one of the repeated portions, and (4) truncating one or more phrases in at least one of the repeated portions; and

    providing the customized speech segment from the server device to a user device for playback with the media asset.

View all claims
  • 1 Assignment

    Thank you for your feedback