Real-Time advanced speech enhancement using low-power, fixed-point hardware

Dariusz Bismor; Krzysztof Bartnicki

doi:10.1109/SPA.2016.7763588

Real-Time advanced speech enhancement using low-power, fixed-point hardware

Dariusz Bismor
, Krzysztof Bartnicki

Department of Measurements and Control Systems (RAU/RAU2)

Oksydan Sp. z o.o.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Citations (Scopus)

Abstract

This paper describes a project aiming at development of an embedded system speech enhancement solution. It was assumed that a low-cost, low-power, fixed-point processor will be used. It was also assumed that both the stationary and nonstationary noises should be removed. To meet the assumptions, the EZDSP5535 evaluation board, with the TMS320C5535 fixed-point digital signal processor, was selected as a hardware platform. Three speech enhancement algorithms were implemented: The FIR filtering, the adaptive line enhancement, and the spectral subtraction. The last algorithm, being a frequencydomain algorithm, utilizes the FFT co-processor available in the TMS320C5535 chip. All the algorithms can work simultaneously. Examples of good final performance are shown.

Original language	English
Title of host publication	2016 Signal Processing
Subtitle of host publication	Algorithms, Architectures, Arrangements, and Applications, SPA 2016
Publisher	IEEE Computer Society
Pages	62-67
Number of pages	6
ISBN (Electronic)	9788362065271
DOIs	https://doi.org/10.1109/SPA.2016.7763588
Publication status	Published - 30 Nov 2016
Event	2016 Signal Processing: Algorithms, Architectures, Arrangements, and Applications, SPA 2016 - Poznan, Poland Duration: 21 Sept 2016 → 23 Sept 2016

Publication series

Name	Signal Processing - Algorithms, Architectures, Arrangements, and Applications Conference Proceedings, SPA
ISSN (Print)	2326-0262
ISSN (Electronic)	2326-0319

Conference

Conference	2016 Signal Processing: Algorithms, Architectures, Arrangements, and Applications, SPA 2016
Country/Territory	Poland
City	Poznan
Period	21/09/16 → 23/09/16

ASJC Scopus subject areas

Computational Theory and Mathematics
Computer Science Applications
Information Systems
Signal Processing

Access to Document

10.1109/SPA.2016.7763588

Cite this

Bismor, D., & Bartnicki, K. (2016). Real-Time advanced speech enhancement using low-power, fixed-point hardware. In 2016 Signal Processing: Algorithms, Architectures, Arrangements, and Applications, SPA 2016 (pp. 62-67). Article 7763588 (Signal Processing - Algorithms, Architectures, Arrangements, and Applications Conference Proceedings, SPA). IEEE Computer Society. https://doi.org/10.1109/SPA.2016.7763588

@inproceedings{4a6570e6e0964e959ca18cc02aa0c527,

title = "Real-Time advanced speech enhancement using low-power, fixed-point hardware",

abstract = "This paper describes a project aiming at development of an embedded system speech enhancement solution. It was assumed that a low-cost, low-power, fixed-point processor will be used. It was also assumed that both the stationary and nonstationary noises should be removed. To meet the assumptions, the EZDSP5535 evaluation board, with the TMS320C5535 fixed-point digital signal processor, was selected as a hardware platform. Three speech enhancement algorithms were implemented: The FIR filtering, the adaptive line enhancement, and the spectral subtraction. The last algorithm, being a frequencydomain algorithm, utilizes the FFT co-processor available in the TMS320C5535 chip. All the algorithms can work simultaneously. Examples of good final performance are shown.",

author = "Dariusz Bismor and Krzysztof Bartnicki",

year = "2016",

month = nov,

day = "30",

doi = "10.1109/SPA.2016.7763588",

language = "English",

series = "Signal Processing - Algorithms, Architectures, Arrangements, and Applications Conference Proceedings, SPA",

publisher = "IEEE Computer Society",

pages = "62--67",

booktitle = "2016 Signal Processing",

address = "United States",

note = "2016 Signal Processing: Algorithms, Architectures, Arrangements, and Applications, SPA 2016 ; Conference date: 21-09-2016 Through 23-09-2016",

}

Bismor, D & Bartnicki, K 2016, Real-Time advanced speech enhancement using low-power, fixed-point hardware. in 2016 Signal Processing: Algorithms, Architectures, Arrangements, and Applications, SPA 2016., 7763588, Signal Processing - Algorithms, Architectures, Arrangements, and Applications Conference Proceedings, SPA, IEEE Computer Society, pp. 62-67, 2016 Signal Processing: Algorithms, Architectures, Arrangements, and Applications, SPA 2016, Poznan, Poland, 21/09/16. https://doi.org/10.1109/SPA.2016.7763588

Real-Time advanced speech enhancement using low-power, fixed-point hardware. / Bismor, Dariusz; Bartnicki, Krzysztof.
2016 Signal Processing: Algorithms, Architectures, Arrangements, and Applications, SPA 2016. IEEE Computer Society, 2016. p. 62-67 7763588 (Signal Processing - Algorithms, Architectures, Arrangements, and Applications Conference Proceedings, SPA).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Real-Time advanced speech enhancement using low-power, fixed-point hardware

AU - Bismor, Dariusz

AU - Bartnicki, Krzysztof

PY - 2016/11/30

Y1 - 2016/11/30

N2 - This paper describes a project aiming at development of an embedded system speech enhancement solution. It was assumed that a low-cost, low-power, fixed-point processor will be used. It was also assumed that both the stationary and nonstationary noises should be removed. To meet the assumptions, the EZDSP5535 evaluation board, with the TMS320C5535 fixed-point digital signal processor, was selected as a hardware platform. Three speech enhancement algorithms were implemented: The FIR filtering, the adaptive line enhancement, and the spectral subtraction. The last algorithm, being a frequencydomain algorithm, utilizes the FFT co-processor available in the TMS320C5535 chip. All the algorithms can work simultaneously. Examples of good final performance are shown.

AB - This paper describes a project aiming at development of an embedded system speech enhancement solution. It was assumed that a low-cost, low-power, fixed-point processor will be used. It was also assumed that both the stationary and nonstationary noises should be removed. To meet the assumptions, the EZDSP5535 evaluation board, with the TMS320C5535 fixed-point digital signal processor, was selected as a hardware platform. Three speech enhancement algorithms were implemented: The FIR filtering, the adaptive line enhancement, and the spectral subtraction. The last algorithm, being a frequencydomain algorithm, utilizes the FFT co-processor available in the TMS320C5535 chip. All the algorithms can work simultaneously. Examples of good final performance are shown.

UR - https://www.scopus.com/pages/publications/85014154877

U2 - 10.1109/SPA.2016.7763588

DO - 10.1109/SPA.2016.7763588

M3 - Conference contribution

AN - SCOPUS:85014154877

T3 - Signal Processing - Algorithms, Architectures, Arrangements, and Applications Conference Proceedings, SPA

SP - 62

EP - 67

BT - 2016 Signal Processing

PB - IEEE Computer Society

T2 - 2016 Signal Processing: Algorithms, Architectures, Arrangements, and Applications, SPA 2016

Y2 - 21 September 2016 through 23 September 2016

ER -

Bismor D, Bartnicki K. Real-Time advanced speech enhancement using low-power, fixed-point hardware. In 2016 Signal Processing: Algorithms, Architectures, Arrangements, and Applications, SPA 2016. IEEE Computer Society. 2016. p. 62-67. 7763588. (Signal Processing - Algorithms, Architectures, Arrangements, and Applications Conference Proceedings, SPA). doi: 10.1109/SPA.2016.7763588

Real-Time advanced speech enhancement using low-power, fixed-point hardware

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this