×

Multi-dimensional disambiguation of voice commands

  • US 8,626,511 B2
  • Filed: 01/22/2010
  • Issued: 01/07/2014
  • Est. Priority Date: 01/22/2010
  • Status: Active Grant
First Claim
Patent Images

1. A computer-implemented method comprising:

  • obtaining two or more candidate transcriptions of a single voice command;

    identifying one or more possible intended actions for each of the two or more candidate transcriptions of the single voice command, including identifying two or more possible intended actions for a particular transcription of the two or more candidate transcriptions of the single voice command;

    providing information for display, the information identifying (i) the two or more candidate transcriptions of the single voice command, and (ii) the one or more possible intended actions for each of the two or more transcriptions of the single voice command, including the two or more possible intended actions for the particular transcription;

    receiving, by one or more computers, data indicating a selection of a particular possible intended action from among the displayed one or more possible intended actions for each of the two or more transcriptions of the single voice command, and the displayed two or more possible intended actions for the particular transcription; and

    invoking the selected particular possible intended action.

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