Iterative joint MAP single-channel speech enhancement given non-uniform phase prior

Pejman Mowlaee Beikzadehmahaleh, Johannes Stahl, Josef Kulmer

Research output: Contribution to journalArticleResearchpeer-review

Abstract

Within the last three decades research in single-channel speech enhancement has been mainly focused on filtering the noisy spectral amplitude without that much focus on the integration of phase-based signal processing. Recently, several phase-aware algorithms based on phase-sensitive signal models were proposed for speech enhancement using the minimum mean squared error (MMSE). Improved performance over the conventional phase-insensitive approaches has been achieved. In this paper, we propose an iterative joint maximum a posteriori (MAP) amplitude and phase estimator (ijMAP) assuming a non-uniform phase distribution. Experimental results demonstrate the effectiveness of the proposed method in recovering both amplitude and phase in noise, justified by perceived quality, speech intelligibility and phase estimation error instrumental measures. The proposed method, brings joint improvement in perceived quality and speech intelligibility compared to the phase-blind joint MAP estimator with a comparable performance to the complex MMSE estimator.
Original languageEnglish
Pages (from-to)85-96
Number of pages12
JournalSpeech Communication
Volume86
DOIs
Publication statusPublished - Feb 2017

Fingerprint

Speech intelligibility
Speech Enhancement
Speech enhancement
Maximum a Posteriori
Error analysis
Signal processing
Speech Intelligibility
Mean Squared Error
performance
Estimator
Phase Error
Enhancement
Error Estimator
Estimation Error
Signal Processing
Filtering

Cite this

Iterative joint MAP single-channel speech enhancement given non-uniform phase prior. / Mowlaee Beikzadehmahaleh, Pejman; Stahl, Johannes; Kulmer, Josef.

In: Speech Communication, Vol. 86, 02.2017, p. 85-96.

Research output: Contribution to journalArticleResearchpeer-review

@article{876352c13029470497dc18e38ab8196c,
title = "Iterative joint MAP single-channel speech enhancement given non-uniform phase prior",
abstract = "Within the last three decades research in single-channel speech enhancement has been mainly focused on filtering the noisy spectral amplitude without that much focus on the integration of phase-based signal processing. Recently, several phase-aware algorithms based on phase-sensitive signal models were proposed for speech enhancement using the minimum mean squared error (MMSE). Improved performance over the conventional phase-insensitive approaches has been achieved. In this paper, we propose an iterative joint maximum a posteriori (MAP) amplitude and phase estimator (ijMAP) assuming a non-uniform phase distribution. Experimental results demonstrate the effectiveness of the proposed method in recovering both amplitude and phase in noise, justified by perceived quality, speech intelligibility and phase estimation error instrumental measures. The proposed method, brings joint improvement in perceived quality and speech intelligibility compared to the phase-blind joint MAP estimator with a comparable performance to the complex MMSE estimator.",
author = "{Mowlaee Beikzadehmahaleh}, Pejman and Johannes Stahl and Josef Kulmer",
year = "2017",
month = "2",
doi = "https://doi.org/10.1016/j.specom.2016.11.008",
language = "English",
volume = "86",
pages = "85--96",
journal = "Speech Communication",
issn = "0167-6393",
publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - Iterative joint MAP single-channel speech enhancement given non-uniform phase prior

AU - Mowlaee Beikzadehmahaleh, Pejman

AU - Stahl, Johannes

AU - Kulmer, Josef

PY - 2017/2

Y1 - 2017/2

N2 - Within the last three decades research in single-channel speech enhancement has been mainly focused on filtering the noisy spectral amplitude without that much focus on the integration of phase-based signal processing. Recently, several phase-aware algorithms based on phase-sensitive signal models were proposed for speech enhancement using the minimum mean squared error (MMSE). Improved performance over the conventional phase-insensitive approaches has been achieved. In this paper, we propose an iterative joint maximum a posteriori (MAP) amplitude and phase estimator (ijMAP) assuming a non-uniform phase distribution. Experimental results demonstrate the effectiveness of the proposed method in recovering both amplitude and phase in noise, justified by perceived quality, speech intelligibility and phase estimation error instrumental measures. The proposed method, brings joint improvement in perceived quality and speech intelligibility compared to the phase-blind joint MAP estimator with a comparable performance to the complex MMSE estimator.

AB - Within the last three decades research in single-channel speech enhancement has been mainly focused on filtering the noisy spectral amplitude without that much focus on the integration of phase-based signal processing. Recently, several phase-aware algorithms based on phase-sensitive signal models were proposed for speech enhancement using the minimum mean squared error (MMSE). Improved performance over the conventional phase-insensitive approaches has been achieved. In this paper, we propose an iterative joint maximum a posteriori (MAP) amplitude and phase estimator (ijMAP) assuming a non-uniform phase distribution. Experimental results demonstrate the effectiveness of the proposed method in recovering both amplitude and phase in noise, justified by perceived quality, speech intelligibility and phase estimation error instrumental measures. The proposed method, brings joint improvement in perceived quality and speech intelligibility compared to the phase-blind joint MAP estimator with a comparable performance to the complex MMSE estimator.

U2 - https://doi.org/10.1016/j.specom.2016.11.008

DO - https://doi.org/10.1016/j.specom.2016.11.008

M3 - Article

VL - 86

SP - 85

EP - 96

JO - Speech Communication

JF - Speech Communication

SN - 0167-6393

ER -