A priori SNR estimation using air- and bone-conduction microphones

Ho Seon Shin, Tim Fingscheidt, Hong-Goo Kang

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

This paper proposes an a priori signal-to-noise ratio (SNR) estimator using an air-conduction (AC) and a bone-conduction (BC) microphone. Among various ways of combining AC and BC microphones for speech enhancement, it is shown that the total enhancement performance can be maximized if the BC microphone is utilized for estimating the power spectral density (PSD) of the desired speech signal. Considering the fact that a small deviation in the speech PSD estimation process brings severe spectral distortion, this paper focuses on controlling weighting factors while estimating the a priori SNR with the decision-directed approach framework. The time-frequency varying weighting factor that is determined by taking a minimum mean square error criterion improves the capability of eliminating residual noise and minimizing speech distortion. Since the weighting factors are also adjusted by measuring the usefulness of the AC and BC microphones, the proposed approach is suitable for tracking the parameter even if the characteristic of environment changes rapidly. The simulation results confirm the superiority of the proposed algorithm to conventional algorithms in high noise environments.

Original languageEnglish
Pages (from-to)2015-2025
Number of pages11
JournalIEEE/ACM Transactions on Audio Speech and Language Processing
Volume23
Issue number11
DOIs
Publication statusPublished - 2015 Nov 1

Fingerprint

Bone Conduction
Signal-To-Noise Ratio
Microphones
microphones
Bone
Conduction
bones
Signal to noise ratio
signal to noise ratios
weighting
Air
air
conduction
Power spectral density
Noise
Weighting
Power Spectral Density
Speech enhancement
Mean square error
estimating

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Media Technology
  • Instrumentation
  • Acoustics and Ultrasonics
  • Linguistics and Language
  • Electrical and Electronic Engineering
  • Speech and Hearing

Cite this

@article{4aa1e33a00444f0bbbfd0dee26bfed37,
title = "A priori SNR estimation using air- and bone-conduction microphones",
abstract = "This paper proposes an a priori signal-to-noise ratio (SNR) estimator using an air-conduction (AC) and a bone-conduction (BC) microphone. Among various ways of combining AC and BC microphones for speech enhancement, it is shown that the total enhancement performance can be maximized if the BC microphone is utilized for estimating the power spectral density (PSD) of the desired speech signal. Considering the fact that a small deviation in the speech PSD estimation process brings severe spectral distortion, this paper focuses on controlling weighting factors while estimating the a priori SNR with the decision-directed approach framework. The time-frequency varying weighting factor that is determined by taking a minimum mean square error criterion improves the capability of eliminating residual noise and minimizing speech distortion. Since the weighting factors are also adjusted by measuring the usefulness of the AC and BC microphones, the proposed approach is suitable for tracking the parameter even if the characteristic of environment changes rapidly. The simulation results confirm the superiority of the proposed algorithm to conventional algorithms in high noise environments.",
author = "Shin, {Ho Seon} and Tim Fingscheidt and Hong-Goo Kang",
year = "2015",
month = "11",
day = "1",
doi = "10.1109/TASLP.2015.2446202",
language = "English",
volume = "23",
pages = "2015--2025",
journal = "IEEE/ACM Transactions on Speech and Language Processing",
issn = "2329-9290",
publisher = "IEEE Advancing Technology for Humanity",
number = "11",

}

A priori SNR estimation using air- and bone-conduction microphones. / Shin, Ho Seon; Fingscheidt, Tim; Kang, Hong-Goo.

In: IEEE/ACM Transactions on Audio Speech and Language Processing, Vol. 23, No. 11, 01.11.2015, p. 2015-2025.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A priori SNR estimation using air- and bone-conduction microphones

AU - Shin, Ho Seon

AU - Fingscheidt, Tim

AU - Kang, Hong-Goo

PY - 2015/11/1

Y1 - 2015/11/1

N2 - This paper proposes an a priori signal-to-noise ratio (SNR) estimator using an air-conduction (AC) and a bone-conduction (BC) microphone. Among various ways of combining AC and BC microphones for speech enhancement, it is shown that the total enhancement performance can be maximized if the BC microphone is utilized for estimating the power spectral density (PSD) of the desired speech signal. Considering the fact that a small deviation in the speech PSD estimation process brings severe spectral distortion, this paper focuses on controlling weighting factors while estimating the a priori SNR with the decision-directed approach framework. The time-frequency varying weighting factor that is determined by taking a minimum mean square error criterion improves the capability of eliminating residual noise and minimizing speech distortion. Since the weighting factors are also adjusted by measuring the usefulness of the AC and BC microphones, the proposed approach is suitable for tracking the parameter even if the characteristic of environment changes rapidly. The simulation results confirm the superiority of the proposed algorithm to conventional algorithms in high noise environments.

AB - This paper proposes an a priori signal-to-noise ratio (SNR) estimator using an air-conduction (AC) and a bone-conduction (BC) microphone. Among various ways of combining AC and BC microphones for speech enhancement, it is shown that the total enhancement performance can be maximized if the BC microphone is utilized for estimating the power spectral density (PSD) of the desired speech signal. Considering the fact that a small deviation in the speech PSD estimation process brings severe spectral distortion, this paper focuses on controlling weighting factors while estimating the a priori SNR with the decision-directed approach framework. The time-frequency varying weighting factor that is determined by taking a minimum mean square error criterion improves the capability of eliminating residual noise and minimizing speech distortion. Since the weighting factors are also adjusted by measuring the usefulness of the AC and BC microphones, the proposed approach is suitable for tracking the parameter even if the characteristic of environment changes rapidly. The simulation results confirm the superiority of the proposed algorithm to conventional algorithms in high noise environments.

UR - http://www.scopus.com/inward/record.url?scp=84942604655&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84942604655&partnerID=8YFLogxK

U2 - 10.1109/TASLP.2015.2446202

DO - 10.1109/TASLP.2015.2446202

M3 - Article

AN - SCOPUS:84942604655

VL - 23

SP - 2015

EP - 2025

JO - IEEE/ACM Transactions on Speech and Language Processing

JF - IEEE/ACM Transactions on Speech and Language Processing

SN - 2329-9290

IS - 11

ER -