Semisupervised autoencoder for sentiment analysis

Z Zhang, S Zhai - US Patent 11,205,103, 2021 - Google Patents
(57) ABSTRACT A method of modelling data, comprising: training an objec tive function of a
linear classifier, based on a set of labeled data, to derive a set of classifier weights; defining …

Multi-party conversation analyzer and logger

JL Gainsboro, LD Weinstein - US Patent 9,300,790, 2016 - Google Patents
(57) ABSTRACT A multi-party conversation analyzer and logger uses a variety of techniques
including spectrographic Voice analysis, abso lute loudness measurements, directional …

Motion tracking system and method using camera and non-camera sensors

E Foxlin, D Wormell, TC Browne… - US Patent …, 2014 - Google Patents
BACKGROUND This description relates to game controllers and 3D input devices for virtual
reality. Console garners have typically interacted with their games through a handheld game …

Combining results from first and second speaker recognition processes

CV AVILÉS-CASCO, MG GOMAR… - US Patent …, 2019 - Google Patents
(57) ABSTRACT A received signal represents a user's speech. A first speaker recognition
process is performed on a first portion of the received signal, to obtain a first output result. A …

Method and apparatus for fraud detection

M Wasserblat, B Eilam, O Pereg, I Kor - US Patent 7,386,105, 2008 - Google Patents
US7386105B2 - Method and apparatus for fraud detection - Google Patents US7386105B2 -
Method and apparatus for fraud detection - Google Patents Method and apparatus for fraud …

Method and apparatus for large population speaker identification in telephone interactions

Y Zigel, M Wasserblat - US Patent 7,822,605, 2010 - Google Patents
(57) ABSTRACT A method and apparatus for determining whether a speaker uttering an
utterance belongs to a predetermined set compris ing known speakers, wherein a training …

Background learning of speaker voices

CS Huang, YC Chu, WH Tsai, JM Cheng - US Patent 7,171,360, 2007 - Google Patents
The invention relates to automatic identification of speak ers by receiving a test utterance;
determining a most likely one of a plurality of speaker models for the test utterance: and …

Speaker verification in a health monitoring system

F Weng, T Hasan, Z Feng - US Patent 8,818,810, 2014 - Google Patents
(57) ABSTRACT A method for Verifying that a person is registered to use a telemedical
device includes identifying an unprompted trig ger phrase in words spoken by a person and …

Channel-compensated low-level features for speaker recognition

E Khoury, M Garland - US Patent 10,347,256, 2019 - Google Patents
ABSTRACT A system for generating channel-compensated features of a speech signal
includes a channel noise simulator that degrades the speech signal, a feed forward …

Text-dependent speaker verification

Z Zhang, A Subramaya - US Patent 8,099,288, 2012 - Google Patents
(57) ABSTRACT A text-dependent speaker verification technique that uses a generic
speaker-independent speech recognizer for robust speaker verification, and uses the …