×

Using speech recognition results based on an unstructured language model in a mobile communication facility application

  • US 8,886,540 B2
  • Filed: 08/01/2008
  • Issued: 11/11/2014
  • Est. Priority Date: 03/07/2007
  • Status: Active Grant
First Claim
Patent Images

1. A method of processing speech, comprising:

  • receiving, at a speech recognition facility, a recording of speech presented by a user using a mobile communication facility resident capture facility, wherein the recording is transmitted through a wireless communication;

    receiving contextual information relating to a software application at the speech recognition facility, wherein the contextual information originated from the mobile communication facility resident capture facility;

    generating results at the speech recognition facility using an unstructured language model based, at least in part, on the contextual information relating to the software application and the recording, wherein the contextual information includes an identity of the mobile communication facility, an identity of a non-speech recognition application resident on the mobile communication facility and a usage history of the non-speech recognition application resident on the mobile communication facility, and wherein user feedback is used to adapt the unstructured language model; and

    transmitting the results to the mobile communications facility, wherein the mobile communication facility is configured to receive the results at the software application and is further configured to simultaneously display the results as a set of words and as a set of application results based on those words.

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