Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

System And Method To Distinguish Between Multiple Logged-on Persons Talking From A Single Node

IP.com Disclosure Number: IPCOM000029193D
Original Publication Date: 2004-Jun-17
Included in the Prior Art Database: 2004-Jun-17
Document File: 2 page(s) / 63K

Publishing Venue

IBM

Abstract

Disclosed is software to distinguish between multiple logged-on persons talking from a single node. In e-meetings, it is important that everyone knows which participant is speaking. The disclosed software will utilize voiceprint and/or camera-based facial recognition technology to determine which of multiple logged-on persons is actively speaking at a particular node.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 1 of 2

System And Method To Distinguish Between Multiple Logged -on Persons Talking From A Single Node

Disclosed is software to distinguish between multiple logged-on persons talking from a single node. In e-meetings, it is important that everyone knows which participant is speaking. The disclosed software will utilize voiceprint and/or camera-based facial recognition technology to determine which of multiple logged-on persons is actively speaking at a particular node.

E-meeting participants use computers and telephones to communicate and manage meeting content. Modern computers have microphone-in connections or support adapter cards with such connections. Even though an e-meeting participant may speak into a telephone, a microphone plugged into his computer will also pick up his speech. Additionally, modern computers have USB and/or FireWire connections for cameras or support adapter cards with such connections. As an e-meeting participant speaks into a telephone, the camera may capture his physical movement. Voice authentication and facial recognition software are well known. They are used to determine one person from another.

In e-meetings, participants are often represented as icons. The computers participating in the e-meeting are nodes. The persons participating in the e-meeting are participants. As nodes connect to the meeting space, they appear as active nodes in the meeting space. The meeting invite list is also in the meeting space. Once a node connects, the user drags the participant or participants from the invite list onto the active node icon. The resulting table, for example, may look like this:

<Sara's Computer>: <TH>Sara

<TH>Bill

<XYZ node>: <TH>Peter

<Unix2837>: <TH>Sam

<TH>Jill <TH>William

<TH> stands for "talking head" icon. This icon is greyed out when the participant is not talking and bright green when the participant is talking.

Both voice and facial recognition software require training before it can be used in an e-meeting. Assume that participants are using the same e-meeting software or software that complies with a voice and facial recognition standard. This assumption allows a successful voice and/or facial training session to be shared across...