Browse Prior Art Database

Voice-Activated Half Duplex Echo Suppressor

IP.com Disclosure Number: IPCOM000040551D
Original Publication Date: 1987-Nov-01
Included in the Prior Art Database: 2005-Feb-02
Document File: 2 page(s) / 26K

Publishing Venue

IBM

Related People

Bustini, L: AUTHOR [+2]

Abstract

The figure shows a schematic representation of a digital voice network. A and B locutors are both connected to the digital network and coders through hybrid transformers (HYB). The hybrid transformers convert each two-wire bidirectional path connected to each telephone set A and B into two unidirectionel paths connected to the digital network. Switches SW1 and SW2 are provided for enabling switching off either one of the unidirectional paths to avoid generation of a feedback or echo signal to either A or B. This is a rough but conventional operation called echo suppressing operation. One of locutors outgoing unidirectional paths is simply turned off based on its relative energy contents.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 59% of the total text.

Page 1 of 2

Voice-Activated Half Duplex Echo Suppressor

The figure shows a schematic representation of a digital voice network. A and B locutors are both connected to the digital network and coders through hybrid transformers (HYB). The hybrid transformers convert each two-wire bidirectional path connected to each telephone set A and B into two unidirectionel paths connected to the digital network. Switches SW1 and SW2 are provided for enabling switching off either one of the unidirectional paths to avoid generation of a feedback or echo signal to either A or B. This is a rough but conventional operation called echo suppressing operation. One of locutors outgoing unidirectional paths is simply turned off based on its relative energy contents. The switching operation is dynamically operated over real time traffic, and quickly repetitive switching operations may occur, deriving into voice chopping. The drawback of this system may be avoided by waiting until the non- switched-off locutor stops talking, rather than monitoring the other locutor's attempts to speak louder to override the switching decision. At t = 0, both SW1 and SW2 are closed. Assuming B starts talking, the signal energy Er becomes greater than E1. The ratio Er/E1 is computed. If it remains greater than a predetermined threshold Alpa during the next two predetermined length time windows, then the system LOGIC considers that B is talking and that the switch on A's side (SW1) should be open to avoid echo return. T...