Real-time voice activity detection for ECoG-based speech brain machine interfaces

Vasileios G. Kanas, Iosif Mporas, Heather L. Benz, Kyriakos N. Sgarbas, Anastasios Bezerianos, Nathan E Crone

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this article, we investigated the performance of a real-time voice activity detection module exploiting different time-frequency methods for extracting signal features in a subject with implanted electrocorticographic (ECoG) electrodes. We used ECoG signals recorded while the subject performed a syllable repetition task. The voice activity detection module used, as input, ECoG data streams, on which it performed feature extraction and classification. With this approach we were able to detect voice activity (speech onset and offset) from ECoG signals with high accuracy. The results demonstrate that different timefrequency representations carried complementary information about voice activity, with the S-transform achieving 92% accuracy using the 86 best features and support vector machines as the classifier. The proposed real-time voice activity detector may be used as a part of an automated natural speech BMI system for rehabilitating individuals with communication deficits.

Original languageEnglish (US)
Title of host publicationInternational Conference on Digital Signal Processing, DSP
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages862-865
Number of pages4
Volume2014-January
ISBN (Print)9781479946129
DOIs
StatePublished - 2014
Event2014 19th International Conference on Digital Signal Processing, DSP 2014 - Hong Kong, Hong Kong
Duration: Aug 20 2014Aug 23 2014

Other

Other2014 19th International Conference on Digital Signal Processing, DSP 2014
CountryHong Kong
CityHong Kong
Period8/20/148/23/14

Fingerprint

Brain
Support vector machines
Feature extraction
Classifiers
Mathematical transformations
Detectors
Electrodes
Communication

Keywords

  • Brain-machine interfaces (BMIs)
  • Electrocorticography (ECoG)
  • Time-frequency analysis
  • Voice activity detection

ASJC Scopus subject areas

  • Signal Processing

Cite this

Kanas, V. G., Mporas, I., Benz, H. L., Sgarbas, K. N., Bezerianos, A., & Crone, N. E. (2014). Real-time voice activity detection for ECoG-based speech brain machine interfaces. In International Conference on Digital Signal Processing, DSP (Vol. 2014-January, pp. 862-865). [6900790] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICDSP.2014.6900790

Real-time voice activity detection for ECoG-based speech brain machine interfaces. / Kanas, Vasileios G.; Mporas, Iosif; Benz, Heather L.; Sgarbas, Kyriakos N.; Bezerianos, Anastasios; Crone, Nathan E.

International Conference on Digital Signal Processing, DSP. Vol. 2014-January Institute of Electrical and Electronics Engineers Inc., 2014. p. 862-865 6900790.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kanas, VG, Mporas, I, Benz, HL, Sgarbas, KN, Bezerianos, A & Crone, NE 2014, Real-time voice activity detection for ECoG-based speech brain machine interfaces. in International Conference on Digital Signal Processing, DSP. vol. 2014-January, 6900790, Institute of Electrical and Electronics Engineers Inc., pp. 862-865, 2014 19th International Conference on Digital Signal Processing, DSP 2014, Hong Kong, Hong Kong, 8/20/14. https://doi.org/10.1109/ICDSP.2014.6900790
Kanas VG, Mporas I, Benz HL, Sgarbas KN, Bezerianos A, Crone NE. Real-time voice activity detection for ECoG-based speech brain machine interfaces. In International Conference on Digital Signal Processing, DSP. Vol. 2014-January. Institute of Electrical and Electronics Engineers Inc. 2014. p. 862-865. 6900790 https://doi.org/10.1109/ICDSP.2014.6900790
Kanas, Vasileios G. ; Mporas, Iosif ; Benz, Heather L. ; Sgarbas, Kyriakos N. ; Bezerianos, Anastasios ; Crone, Nathan E. / Real-time voice activity detection for ECoG-based speech brain machine interfaces. International Conference on Digital Signal Processing, DSP. Vol. 2014-January Institute of Electrical and Electronics Engineers Inc., 2014. pp. 862-865
@inproceedings{18d8de1171164b30ae7d8770acf0cf7b,
title = "Real-time voice activity detection for ECoG-based speech brain machine interfaces",
abstract = "In this article, we investigated the performance of a real-time voice activity detection module exploiting different time-frequency methods for extracting signal features in a subject with implanted electrocorticographic (ECoG) electrodes. We used ECoG signals recorded while the subject performed a syllable repetition task. The voice activity detection module used, as input, ECoG data streams, on which it performed feature extraction and classification. With this approach we were able to detect voice activity (speech onset and offset) from ECoG signals with high accuracy. The results demonstrate that different timefrequency representations carried complementary information about voice activity, with the S-transform achieving 92{\%} accuracy using the 86 best features and support vector machines as the classifier. The proposed real-time voice activity detector may be used as a part of an automated natural speech BMI system for rehabilitating individuals with communication deficits.",
keywords = "Brain-machine interfaces (BMIs), Electrocorticography (ECoG), Time-frequency analysis, Voice activity detection",
author = "Kanas, {Vasileios G.} and Iosif Mporas and Benz, {Heather L.} and Sgarbas, {Kyriakos N.} and Anastasios Bezerianos and Crone, {Nathan E}",
year = "2014",
doi = "10.1109/ICDSP.2014.6900790",
language = "English (US)",
isbn = "9781479946129",
volume = "2014-January",
pages = "862--865",
booktitle = "International Conference on Digital Signal Processing, DSP",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Real-time voice activity detection for ECoG-based speech brain machine interfaces

AU - Kanas, Vasileios G.

AU - Mporas, Iosif

AU - Benz, Heather L.

AU - Sgarbas, Kyriakos N.

AU - Bezerianos, Anastasios

AU - Crone, Nathan E

PY - 2014

Y1 - 2014

N2 - In this article, we investigated the performance of a real-time voice activity detection module exploiting different time-frequency methods for extracting signal features in a subject with implanted electrocorticographic (ECoG) electrodes. We used ECoG signals recorded while the subject performed a syllable repetition task. The voice activity detection module used, as input, ECoG data streams, on which it performed feature extraction and classification. With this approach we were able to detect voice activity (speech onset and offset) from ECoG signals with high accuracy. The results demonstrate that different timefrequency representations carried complementary information about voice activity, with the S-transform achieving 92% accuracy using the 86 best features and support vector machines as the classifier. The proposed real-time voice activity detector may be used as a part of an automated natural speech BMI system for rehabilitating individuals with communication deficits.

AB - In this article, we investigated the performance of a real-time voice activity detection module exploiting different time-frequency methods for extracting signal features in a subject with implanted electrocorticographic (ECoG) electrodes. We used ECoG signals recorded while the subject performed a syllable repetition task. The voice activity detection module used, as input, ECoG data streams, on which it performed feature extraction and classification. With this approach we were able to detect voice activity (speech onset and offset) from ECoG signals with high accuracy. The results demonstrate that different timefrequency representations carried complementary information about voice activity, with the S-transform achieving 92% accuracy using the 86 best features and support vector machines as the classifier. The proposed real-time voice activity detector may be used as a part of an automated natural speech BMI system for rehabilitating individuals with communication deficits.

KW - Brain-machine interfaces (BMIs)

KW - Electrocorticography (ECoG)

KW - Time-frequency analysis

KW - Voice activity detection

UR - http://www.scopus.com/inward/record.url?scp=84940747522&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84940747522&partnerID=8YFLogxK

U2 - 10.1109/ICDSP.2014.6900790

DO - 10.1109/ICDSP.2014.6900790

M3 - Conference contribution

AN - SCOPUS:84940747522

SN - 9781479946129

VL - 2014-January

SP - 862

EP - 865

BT - International Conference on Digital Signal Processing, DSP

PB - Institute of Electrical and Electronics Engineers Inc.

ER -