Postfiltering Using Multichannel Spectral Estimation in Multispeaker Environments

This paper investigates the problem of enhancing a single desired speech source from a mixture of signals in multispeaker environments. A beamformer structure is proposed which combines a fixed beamformer with postfiltering. In the first stage, the fixed multiobjective optimal beamformer is designed...

Full description

Bibliographic Details
Main Authors: Dam, Hai, Nordholm, Sven, Dam, Hai Huyen Heidi, Low, Siow Yong
Format: Journal Article
Published: Hindawi Publishing Corporation 2008
Online Access:http://hdl.handle.net/20.500.11937/3696
_version_ 1848744302258159616
author Dam, Hai
Nordholm, Sven
Dam, Hai Huyen Heidi
Low, Siow Yong
author_facet Dam, Hai
Nordholm, Sven
Dam, Hai Huyen Heidi
Low, Siow Yong
author_sort Dam, Hai
building Curtin Institutional Repository
collection Online Access
description This paper investigates the problem of enhancing a single desired speech source from a mixture of signals in multispeaker environments. A beamformer structure is proposed which combines a fixed beamformer with postfiltering. In the first stage, the fixed multiobjective optimal beamformer is designed to spatially extract the desired source by suppressing all other undesired sources. In the second stage, a multichannel power spectral estimator is proposed and incorporated in the postfilter, thus enabling further suppression capability. The combined scheme exploits both spatial and spectral characteristics of the signals. Two new multichannel spectral estimation methods are proposed for the postfiltering using, respectively, inner product and joint diagonalization. Evaluations using recordings from a real-room environment show that the proposed beamformer offers a good interference suppression level whilst maintaining a low-distortion level of the desired source.
first_indexed 2025-11-14T05:59:18Z
format Journal Article
id curtin-20.500.11937-3696
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T05:59:18Z
publishDate 2008
publisher Hindawi Publishing Corporation
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-36962017-09-13T14:43:26Z Postfiltering Using Multichannel Spectral Estimation in Multispeaker Environments Dam, Hai Nordholm, Sven Dam, Hai Huyen Heidi Low, Siow Yong This paper investigates the problem of enhancing a single desired speech source from a mixture of signals in multispeaker environments. A beamformer structure is proposed which combines a fixed beamformer with postfiltering. In the first stage, the fixed multiobjective optimal beamformer is designed to spatially extract the desired source by suppressing all other undesired sources. In the second stage, a multichannel power spectral estimator is proposed and incorporated in the postfilter, thus enabling further suppression capability. The combined scheme exploits both spatial and spectral characteristics of the signals. Two new multichannel spectral estimation methods are proposed for the postfiltering using, respectively, inner product and joint diagonalization. Evaluations using recordings from a real-room environment show that the proposed beamformer offers a good interference suppression level whilst maintaining a low-distortion level of the desired source. 2008 Journal Article http://hdl.handle.net/20.500.11937/3696 10.1155/2008/860360 Hindawi Publishing Corporation fulltext
spellingShingle Dam, Hai
Nordholm, Sven
Dam, Hai Huyen Heidi
Low, Siow Yong
Postfiltering Using Multichannel Spectral Estimation in Multispeaker Environments
title Postfiltering Using Multichannel Spectral Estimation in Multispeaker Environments
title_full Postfiltering Using Multichannel Spectral Estimation in Multispeaker Environments
title_fullStr Postfiltering Using Multichannel Spectral Estimation in Multispeaker Environments
title_full_unstemmed Postfiltering Using Multichannel Spectral Estimation in Multispeaker Environments
title_short Postfiltering Using Multichannel Spectral Estimation in Multispeaker Environments
title_sort postfiltering using multichannel spectral estimation in multispeaker environments
url http://hdl.handle.net/20.500.11937/3696