×

Method for the voice recognition of a speaker using a predictive model, particularly for access control applications

  • US 6,349,279 B1
  • Filed: 04/05/1999
  • Issued: 02/19/2002
  • Est. Priority Date: 05/03/1996
  • Status: Expired due to Fees
First Claim
Patent Images

1. A method for the voice recognition of a speaker using a q-order predictive model comprising at least one phase of extracting statistical characteristics (P4, P′

  • 4) including at least one step of digital acquisition of a voice sample of a particular duration D of the speaker (U) corresponding to at least one utterance of the speaker (U), a step of converting said voice sample into a sequence of vectors of particular size p obtained from a sequence of analysis windows of average duration T and with average spacing I, and a step of determining q+1 correlation matrices of size p×

    p from this sequence of vectors, where p and q are non-zero integers, characterized in that said average duration T has a duration less than 10 ms.

View all claims
  • 1 Assignment
Timeline View
Assignment View
    ×
    ×