×

Automated speech pronunciation attribution

  • US 10,559,296 B2
  • Filed: 06/01/2018
  • Issued: 02/11/2020
  • Est. Priority Date: 12/29/2016
  • Status: Active Grant
First Claim
Patent Images

1. A computer-implemented method comprising:

  • receiving, by a digital assistant device that stores multiple user profiles that are each associated with a respective one of multiple users, a voice command of a particular one of the multiple users, wherein the voice command includes a particular term that, among the multiple users, is pronounced uniquely by the particular one of the multiple users, and wherein each user profile stored by the digital assistant device specifies pronunciation data for terms that the respective user pronounces uniquely;

    matching the voice command to a particular user profile among the multiple stored user profiles that are stored by the digital assistant;

    generating, by the digital assistant device, an acknowledgment of the voice command, wherein the acknowledgement includes the particular term and pronunciation data that was stored in the matched, particular user profile and that reflects the unique pronunciation of the particular term by the particular one of the multiple users; and

    providing, for output by a speech synthesizer of the digital assistant device, a spoken representation of the acknowledgment, wherein the spoken representation of the acknowledgement of the voice command includes the particular term as uniquely pronounced by the particular one of the multiple users.

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