Abstract

Our goal is to study the statistical methods for source separation based on temporal and frequency specific features by using particle filtering. Particle filtering is an advanced state-space Bayesian estimation technique that supports non-Gaussian and nonlinear models along with time-varying noise, allowing for a more accurate model of the underlying system dynamics. We present a system that combines standard speech processing techniques in a novel method to separate two noisy speech sources. The system models the pitch and amplitude over time separately, and adopts particle filtering to reduce complexity by generating a discrete distribution that approximates well the desired continuous distribution. Preliminary results that demonstrate the separation of two noisy sources using this system are presented.

Original languageEnglish (US)
Pages2673-2676
Number of pages4
StatePublished - Jan 1 2004
Event8th International Conference on Spoken Language Processing, ICSLP 2004 - Jeju, Jeju Island, Korea, Republic of
Duration: Oct 4 2004Oct 8 2004

Other

Other8th International Conference on Spoken Language Processing, ICSLP 2004
CountryKorea, Republic of
CityJeju, Jeju Island
Period10/4/0410/8/04

Fingerprint

non-linear model
estimation procedure
system model
statistical method
Filter
Particle
time
Speech Processing
Statistical Methods
Dynamic Systems

ASJC Scopus subject areas

  • Language and Linguistics
  • Linguistics and Language

Cite this

Gandhi, M. A., & Hasegawa-Johnson, M. A. (2004). Source separation using particle filters. 2673-2676. Paper presented at 8th International Conference on Spoken Language Processing, ICSLP 2004, Jeju, Jeju Island, Korea, Republic of.

Source separation using particle filters. / Gandhi, Mital A.; Hasegawa-Johnson, Mark Allan.

2004. 2673-2676 Paper presented at 8th International Conference on Spoken Language Processing, ICSLP 2004, Jeju, Jeju Island, Korea, Republic of.

Research output: Contribution to conferencePaper

Gandhi, MA & Hasegawa-Johnson, MA 2004, 'Source separation using particle filters', Paper presented at 8th International Conference on Spoken Language Processing, ICSLP 2004, Jeju, Jeju Island, Korea, Republic of, 10/4/04 - 10/8/04 pp. 2673-2676.
Gandhi MA, Hasegawa-Johnson MA. Source separation using particle filters. 2004. Paper presented at 8th International Conference on Spoken Language Processing, ICSLP 2004, Jeju, Jeju Island, Korea, Republic of.
Gandhi, Mital A. ; Hasegawa-Johnson, Mark Allan. / Source separation using particle filters. Paper presented at 8th International Conference on Spoken Language Processing, ICSLP 2004, Jeju, Jeju Island, Korea, Republic of.4 p.
@conference{3b6f61836d114019b111f1bd85d08b05,
title = "Source separation using particle filters",
abstract = "Our goal is to study the statistical methods for source separation based on temporal and frequency specific features by using particle filtering. Particle filtering is an advanced state-space Bayesian estimation technique that supports non-Gaussian and nonlinear models along with time-varying noise, allowing for a more accurate model of the underlying system dynamics. We present a system that combines standard speech processing techniques in a novel method to separate two noisy speech sources. The system models the pitch and amplitude over time separately, and adopts particle filtering to reduce complexity by generating a discrete distribution that approximates well the desired continuous distribution. Preliminary results that demonstrate the separation of two noisy sources using this system are presented.",
author = "Gandhi, {Mital A.} and Hasegawa-Johnson, {Mark Allan}",
year = "2004",
month = "1",
day = "1",
language = "English (US)",
pages = "2673--2676",
note = "8th International Conference on Spoken Language Processing, ICSLP 2004 ; Conference date: 04-10-2004 Through 08-10-2004",

}

TY - CONF

T1 - Source separation using particle filters

AU - Gandhi, Mital A.

AU - Hasegawa-Johnson, Mark Allan

PY - 2004/1/1

Y1 - 2004/1/1

N2 - Our goal is to study the statistical methods for source separation based on temporal and frequency specific features by using particle filtering. Particle filtering is an advanced state-space Bayesian estimation technique that supports non-Gaussian and nonlinear models along with time-varying noise, allowing for a more accurate model of the underlying system dynamics. We present a system that combines standard speech processing techniques in a novel method to separate two noisy speech sources. The system models the pitch and amplitude over time separately, and adopts particle filtering to reduce complexity by generating a discrete distribution that approximates well the desired continuous distribution. Preliminary results that demonstrate the separation of two noisy sources using this system are presented.

AB - Our goal is to study the statistical methods for source separation based on temporal and frequency specific features by using particle filtering. Particle filtering is an advanced state-space Bayesian estimation technique that supports non-Gaussian and nonlinear models along with time-varying noise, allowing for a more accurate model of the underlying system dynamics. We present a system that combines standard speech processing techniques in a novel method to separate two noisy speech sources. The system models the pitch and amplitude over time separately, and adopts particle filtering to reduce complexity by generating a discrete distribution that approximates well the desired continuous distribution. Preliminary results that demonstrate the separation of two noisy sources using this system are presented.

UR - http://www.scopus.com/inward/record.url?scp=64149118623&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=64149118623&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:64149118623

SP - 2673

EP - 2676

ER -