×

SPEECH RECOGNITION SYSTEM AND METHOD

  • US 20110029316A1
  • Filed: 10/18/2010
  • Published: 02/03/2011
  • Est. Priority Date: 12/23/1998
  • Status: Active Grant
First Claim
Patent Images

1. A computer readable storage medium encoded with a plurality of instructions that, when executed by a computer, perform a method for integrating processes within a voice command human interface, the method comprising:

  • receiving a voice signal comprising a contextually defined command sequence;

    determining a context;

    processing the voice signal using a context-dependent natural language model;

    parsing executable commands from the processed voice signal;

    determining a set of processes appropriate for execution of the parsed commands;

    routing the parsed commands to the appropriate processes for execution; and

    executing the contextually defined command sequence, wherein a plurality of processes execute concurrently.

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