IP Library Granted Patent US 7,672,847
Granted Patent B2
US 7,672,847 · App. 12/241,811 · Granted Mar 2, 2010

Discriminative training of hidden Markov models for continuous speech recognition

View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 7,672,847
App. No.
12/241,811
Granted
Mar 2, 2010
Kind
B2
Abstract

Methods are given for improving discriminative training of hidden Markov models for continuous speech recognition. For a mixture component of a hidden Markov model state, a gradient adjustment is calculated of the standard deviation of the mixture component. If the calculated gradient adjustment is greater than a first threshold amount, an adjustment is performed of the standard deviation of the mixture component using the first threshold. If the calculated gradient adjustment is less than a second threshold amount, an adjustment is performed of the standard deviation of the mixture component using the second threshold. Otherwise, an adjustment is performed of the standard deviation of the mixture component using the calculated gradient adjustment.

Claims (5)

1. A method of a continuous speech recognition system for discriminatively training hidden Markov models, the method comprising:

for a mixture component of a hidden Markov model state, calculating in a speech recognition system a gradient adjustment of the standard deviation of the mixture component, and

i. if the calculated gradient adjustment is greater than a first threshold amount, performing an adjustment of the standard deviation of the mixture component using the first threshold, or

ii. if the calculated gradient adjustment is less than a second threshold amount, performing an adjustment of the standard deviation of the mixture component using the second threshold, else

iii. performing an adjustment of the standard deviation of the mixture component using the calculated gradient adjustment to produce an adjusted hidden Markov model for speech recognition.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 3, 2008
From: HE, CHUANG; WU, JIANXIONG; SEJNOHA, VLAD
To: SCANSOFT, INC.
Reel/Frame 021631/0939 →
MERGER AND CHANGE OF NAME TO NUANCE COMMUNICATIONS, INC. Recorded Oct 3, 2008
From: SCANSOFT, INC.
To: NUANCE COMMUNICATIONS, INC.
Reel/Frame 021633/0207 →