Patent · US Active

Differential acoustic model representation and linear transform-based adaptation for efficient user profile update techniques in automatic speech recognition

US9406299B2 · kind B2 · utility

138Cited by

5References

20Claims

0Family size

Assignee

Nuance Communications, Inc. · US

Inventors

Christian Gollan · Aachen, DE
Daniel Willett · Eltville am Rhein, DE

Key dates

Filing date	Mar 8, 2012
Grant date	Aug 2, 2016
Priority date	—
Expiry date	Apr 23, 2032

Classification

Technology area (CPC G)Physics
CPC primaryG10L2015/227
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

A computer-implemented method is described for speaker adaptation in automatic speech recognition. Speech recognition data from a particular speaker is used for adaptation of an initial speech recognition acoustic model to produce a speaker adapted acoustic model. A speaker dependent differential acoustic model is determined that represents differences between the initial speech recognition acoustic model and the speaker adapted acoustic model. In addition, an approach is also disclosed to estimate speaker-specific feature or model transforms over multiple sessions. This is achieved by updating the previously estimated transform using only adaptation statistics of the current session.

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