Browse Prior Art Database

Technique for detection of a person in a conference call, to detect a user action requiring unmute/mute, and to do it automatically Disclosure Number: IPCOM000236190D
Publication Date: 2014-Apr-11
Document File: 3 page(s) / 58K

Publishing Venue

The Prior Art Database


Technique to automatically mute/unmute user on a conference call if he uses native or non-preferrred language

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 3

Technique for detection of a person in a conference call , to detect a user action requiring unmute/mute, and to do it automatically

Conference calls are very common for most of the working class where users regularly dial into conference calls and put their phone on mute or unmute depending on whether they want to speak. Although there exist methods to automatically mute/unmute based on certain conditions, here, we propose a the following approach:

Technique to filter all the background noise in a conference as the user speaks (in a crowded place).

Technique to detect when the user is speaking in NATIVE (non-preferred) language and MUTE the phone, and alternatively when speaking in the "Preferred" language of the conference, UNMUTE the phone
Technique to detect when the user has completed the conversation and MUTE the phone


The solution is to be implemented at the hand-set rather than at the server side. This is because the MUTE and UNMUTE are being handled at the phone end. The solution can also be implemented at the conference server end which then does not require specific features inside the phone which is then always unmuted, but the user is muted by the server. So this can be useful for handset OS'es such as Android/iOS as well as to conference server makers like WebEx.

Intelligence for detecting when the user is dialing into a conference call ->

There already exist ways to detect this. e.g.Rules engine to detect when the user is in a conference call. This could be as simple as a UI element in the phone number which user stores, or more intelligent by scanning the carrier or the web for the phone number to be dialed. For example, if you search on web for 000117 airtel, you can clearly see that it's a toll-free number to log into AT&T conference system. Similarly searching on 888426640 gives similar information. So this can be easily achieved with intelligence & learning. Another use-case is when the user has dialed into a single call (1 to 1 ), but then the other party conferences in another user or users. So this boils down to detecting a switchover from 1-1 to 1-many call.

Rules engine which defines rules for UNMUTE of the phone There can be multiple ways of doing this:

Voice/Noise on the user...