Patent · US Active

Method and apparatus for detecting spoofing conditions

US10692502B2 · kind B2 · utility

5Cited by

8References

22Claims

0Family size

Assignee

Pindrop Security, Inc. · US

Inventors

Elie Khoury · Atlanta, US
Parav Nagarsheth · Atlanta, US
Kailash Patil · Atlanta, US
Matthew Garland · Atlanta, US

Key dates

Filing date	Mar 2, 2018
Grant date	Jun 23, 2020
Priority date	—
Expiry date	Jun 5, 2038

Classification

Technology area (CPC G)Physics
CPC primaryG10L25/51
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

An automated speaker verification (ASV) system incorporates a first deep neural network to extract deep acoustic features, such as deep CQCC features, from a received voice sample. The deep acoustic features are processed by a second deep neural network that classifies the deep acoustic features according to a determined likelihood of including a spoofing condition. A binary classifier then classifies the voice sample as being genuine or spoofed.

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