Patent · US Active

Source signal separation by discriminatively-trained non-negative matrix factorization

US9679559B2 · kind B2 · utility

2Cited by

0References

7Claims

0Family size

Assignee

Mitsubishi Electric Research Laboratories, Inc. · US

Inventors

Jonathan Le Roux · Somerville, US
John R. Hershey · Winchester, US
Felix Weninger · Munich, DE
Shinji Watanabe · Minato, JP

Key dates

Filing date	May 29, 2014
Grant date	Jun 13, 2017
Priority date	—
Expiry date	May 29, 2034

Classification

Technology area (CPC G)Physics
CPC primaryG10L21/0232
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

A method estimates source signals from a mixture of source signals by first training an analysis model and a reconstruction model using training data. The analysis model is applied to the mixture of source signals to obtain an analysis representation of the mixture of source signals, and the reconstruction model is applied to the analysis representation to obtain an estimate of the source signals, wherein the analysis model utilizes an analysis linear basis representation, and the reconstruction model utilizes a reconstruction linear basis representation.

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