Using multiple attributes to create a voice search playlist

  • US 7,801,729 B2
  • Filed: 03/13/2007
  • Issued: 09/21/2010
  • Est. Priority Date: 03/13/2007
  • Status: Active Grant
  • ×
    • Pin Icon | RPX Insight
    • Pin
First Claim
Patent Images

1. A speech recognition method comprising:

  • storing a first plurality of content on an electronic device, wherein each item of content is associated with a second plurality of content attributes, each content attribute having a content attribute value;

    adding the content attribute values for the second plurality of content attributes for the first plurality of content to a first recognition set of a speech recognizer;

    receiving a speech input signal in said speech recognizer;

    generating a plurality of likelihood values in response to the speech input signal, wherein each likelihood value is associated with one content attribute value in the recognition set; and

    accessing the stored content based on the likelihood values and generating a play list based on the likelihood values,wherein, for a third plurality of content attribute values having an associated likelihood value greater than a threshold, content associated with said third plurality of content attribute values is included in the play list.

View all claims

    Thank you for your feedback