Patent · US Active

Speech models generated using competitive training, asymmetric training, and data boosting

US8532991B2 · kind B2 · utility

2Cited by
9References
20Claims
0Family size

Assignee

Inventors

Key dates

Filing dateMar 10, 2010
Grant dateSep 10, 2013
Priority date
Expiry dateMar 5, 2031

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG10L15/063
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

Speech models are trained using one or more of three different training systems. They include competitive training which reduces a distance between a recognized result and a true result, data boosting which divides and weights training data, and asymmetric training which trains different model components differently.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.