Computational Complexity Reduction with Mel-Frequency Filterbank-Based Approach for Multichannel Speech Enhancement
Summary :
Multichannel speech enhancement systems (MSES') have been widely utilized for diverse types of speech interface applications. A state-of-the-art MSES primarily utilizes multichannel minima-controlled recursive averaging for noise estimations and a parameterized multichannel Wiener filter for noise reduction. Many MSES' are implemented in the frequency domain, but they are computationally burdensome due to the numerous complex matrix operations involved. In this paper, a novel MSES intended to reduce the computational complexity with improved performance is proposed. The proposed system is implemented in the mel-filterbank domain using a frequency-averaging technique. Through a performance evaluation, it is verified that the proposed mel-filterbank MSES achieves improvements in the perceptual speech quality with a reduced level of computation compared to a conventional MSES.
- Publication
- IEICE TRANSACTIONS on Fundamentals Vol.E100-A No.10 pp.2154-2157
- Publication Date
- 2017/10/01
- Publicized
- Online ISSN
- 1745-1337
- DOI
- 10.1587/transfun.E100.A.2154
- Type of Manuscript
- LETTER
- Category
- Speech and Hearing
Authors
Jungpyo HONG
KAIST
Sangbae JEONG
Gyeongsang Natl. Univ.
Keyword
Latest Issue
Copyrights notice of machine-translated contents
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Cite this
Copy
Jungpyo HONG, Sangbae JEONG, "Computational Complexity Reduction with Mel-Frequency Filterbank-Based Approach for Multichannel Speech Enhancement" in IEICE TRANSACTIONS on Fundamentals,
vol. E100-A, no. 10, pp. 2154-2157, October 2017, doi: 10.1587/transfun.E100.A.2154.
Abstract: Multichannel speech enhancement systems (MSES') have been widely utilized for diverse types of speech interface applications. A state-of-the-art MSES primarily utilizes multichannel minima-controlled recursive averaging for noise estimations and a parameterized multichannel Wiener filter for noise reduction. Many MSES' are implemented in the frequency domain, but they are computationally burdensome due to the numerous complex matrix operations involved. In this paper, a novel MSES intended to reduce the computational complexity with improved performance is proposed. The proposed system is implemented in the mel-filterbank domain using a frequency-averaging technique. Through a performance evaluation, it is verified that the proposed mel-filterbank MSES achieves improvements in the perceptual speech quality with a reduced level of computation compared to a conventional MSES.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.E100.A.2154/_p
Copy
@ARTICLE{e100-a_10_2154,
author={Jungpyo HONG, Sangbae JEONG, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Computational Complexity Reduction with Mel-Frequency Filterbank-Based Approach for Multichannel Speech Enhancement},
year={2017},
volume={E100-A},
number={10},
pages={2154-2157},
abstract={Multichannel speech enhancement systems (MSES') have been widely utilized for diverse types of speech interface applications. A state-of-the-art MSES primarily utilizes multichannel minima-controlled recursive averaging for noise estimations and a parameterized multichannel Wiener filter for noise reduction. Many MSES' are implemented in the frequency domain, but they are computationally burdensome due to the numerous complex matrix operations involved. In this paper, a novel MSES intended to reduce the computational complexity with improved performance is proposed. The proposed system is implemented in the mel-filterbank domain using a frequency-averaging technique. Through a performance evaluation, it is verified that the proposed mel-filterbank MSES achieves improvements in the perceptual speech quality with a reduced level of computation compared to a conventional MSES.},
keywords={},
doi={10.1587/transfun.E100.A.2154},
ISSN={1745-1337},
month={October},}
Copy
TY - JOUR
TI - Computational Complexity Reduction with Mel-Frequency Filterbank-Based Approach for Multichannel Speech Enhancement
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 2154
EP - 2157
AU - Jungpyo HONG
AU - Sangbae JEONG
PY - 2017
DO - 10.1587/transfun.E100.A.2154
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E100-A
IS - 10
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - October 2017
AB - Multichannel speech enhancement systems (MSES') have been widely utilized for diverse types of speech interface applications. A state-of-the-art MSES primarily utilizes multichannel minima-controlled recursive averaging for noise estimations and a parameterized multichannel Wiener filter for noise reduction. Many MSES' are implemented in the frequency domain, but they are computationally burdensome due to the numerous complex matrix operations involved. In this paper, a novel MSES intended to reduce the computational complexity with improved performance is proposed. The proposed system is implemented in the mel-filterbank domain using a frequency-averaging technique. Through a performance evaluation, it is verified that the proposed mel-filterbank MSES achieves improvements in the perceptual speech quality with a reduced level of computation compared to a conventional MSES.
ER -
English
