Multi-Modal Acoustic Echo Canceller for Video Conferencing Systems

In video conferencing, if two people talk at the same time howling noises can appear due to failures in the echo cancellation system. In order to reduce these noise situations, a novel AEC (Acoustic Echo Cancellation) system is proposed by using audio and video information. Through image processing...

Full description

Saved in:
Bibliographic Details
Main Authors: Gazziro, M., Almeida, G., Matias, P., Tanaka, H., Minami, S.
Format: Conference Proceeding
Language:English
Subjects:
Acoustic Echo Cancellation
Adaptive filters
Algorithm design and analysis
Detectors
Double Talk Detection
Echo cancellers
Face
Image Processing
Mouth
Multi-Modal
Video Conference
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In video conferencing, if two people talk at the same time howling noises can appear due to failures in the echo cancellation system. In order to reduce these noise situations, a novel AEC (Acoustic Echo Cancellation) system is proposed by using audio and video information. Through image processing techniques we have developed a new DTD (Double Talk Detector) based on video information (video-DTD). A novel Acoustic Echo Cancellation system was successfully developed, integrating audio and video information. In simulated results using data from a real scenary the misalignment of the new AEC (with video-DTD) was smaller than a traditional AEC.
ISSN:1530-1834
2377-5416
DOI:10.1109/SIBGRAPI.2012.49