Vincent W. Neo

Vincent W. Neo

I am a postdoctoral researcher in the Speech and Audio Processing Lab at Imperial College London, UK

Publications

  1. Polynomial Eigenvalue Decomposition for Multichannel Broadband Signal Processing

    Published in IEEE Signal Processing Magazine, 2023

    This paper introduces polynomial matrix eigenvalue decomposition (PEVD) and applications to multichannel broadband signal processing.

    Recommended citation: V. W. Neo, S. Redif, J. G. McWhirter, J. Pestana, I. K. Proudler, S. Weiss, and P. A. Naylor, "Polynomial Eigenvalue Decomposition for Multichannel Broadband Signal Processing", IEEE Signal Process. Mag., vol. 40, no. 7, 2023. doi: 10.1109/MSP.2023.3269200

  2. Signal Compaction Using PEVD for Spherical Array Processing with Applications

    Published in IEEE/ACM Trans. Audio Speech and Lang. Process., 2023

    This paper presents the a unified framework using SHT and PEVD for applications using spherical sensor arrays.

    Recommended citation: V. W. Neo, C. Evers, S.Weiss, and P. A. Naylor, "Signal compaction using PEVD for spherical array processing with applications," IEEE/ACM Trans. Audio, Speech and Lang. Process., vol. 28, 2023. doi: 10.1109/TASLP.2023.3313441

  3. Dual Input Neural Networks for Positional Sound Source Localization

    Published in EURASIP Journal on Audio, Speech, and Music Processing, 2023

    This paper proposes new neural network approaches to jointly process signals and exploit metadata. The effectiveness of these approaches are demonstrated for the positional sound source localization task.

    Recommended citation: E. Grinstein, V. W. Neo, and P. A. Naylor, "Dual input neural networks for positional sound source localization," EURASIP J. Audio, Speech, and, Music Process., vol. 32, Aug. 2023, pp. 1-12, https://doi.org/10.1186/s13636-023-00301-x

  4. Studying Human-Based Speaker Diarization and Comparing to State-of-the-Art Systems

    Published in ASPIPA, 2022

    This paper compares the performance of human-based speaker diarization against state-of-the-art machine learning systems.

    Recommended citation: S. W. McKnight, A. O. T. Hogg, V. W. Neo, and P. A. Naylor, "Studying human-based speaker diarization and comparing to state-of-the-art systems," in Proc. Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC), Nov. 2022.

  5. Fixed Beamformer Design Using Polynomial Eigenvalue Decomposition

    Published in IWAENC, 2022

    This paper presents a fixed beamformer design using PEVD.

    Recommended citation: V. W. Neo, E. d'Olne, A. H. Moore, and P. A. Naylor, "Fixed beamformer design using polynomial eigenvalue decomposition," in Proc. Int. Workshop on Acoust. Signal Enhancement (IWAENC), Sep. 2022.

  6. Frame-based Space-time Covariance Matrix Estimation for Polynomial Eigenvalue Decomposition-based Speech Enhancement

    Published in IWAENC, 2022

    This paper presents a frame-based space-time covariance matrix estimation for PEVD-based speech enhancement.

    Recommended citation: E. d'Olne, V. W. Neo, and P. A. Naylor, "Frame-based space-time covariance matrix estimation for polynomial eigenvalue decomposition-based speech enhancement," in Proc. Int. Workshop on Acoust. Signal Enhancement (IWAENC), Sep. 2022.

  7. Enhancement of Noisy Reverberant Speech Using Polynomial Matrix Eigenvalue Decomposition

    Published in IEEE/ACM Trans. Audio Speech and Lang. Process., 2021

    This paper presents the PEVD-based speech enhancement algorithm under a unified framework.

    Recommended citation: V. W. Neo, C. Evers, and P. A. Naylor, "Enhancement of noisy reverberant speech using polynomial matrix eigenvalue decomposition," IEEE/ACM Trans. Audio, Speech and Lang. Process., vol. 28, 2021. doi: 10.1109/TASLP.2021.3120630

  8. A Study of Salient Modulation Domain Features for Speaker Identification

    Published in ASPIPA, 2021

    This paper studies the modulation domain features useful for speaker identification.

    Recommended citation: S. W. McKnight, A. O. T. Hogg, V. W. Neo, and P. A. Naylor, "A study of salient modulation domain features for speaker identification," in Proc. Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC), Dec. 2021.

  9. A Polynomial Eigenvalue Decomposition MUSIC Approach for Broadband Sound Source Localization

    Published in WASPAA, 2021

    This paper proposes enhancements to polynomial MUSIC (PMUSIC), a broadband extension to the classical MUSIC algorithm for sound source localization.

    Recommended citation: A. O. T. Hogg, V. W. Neo, C. Evers and P. A. Naylor, "A polynomial eigenvalue decomposition MUSIC approach for broadband sound source localization," in Proc. IEEE Workshop on Applications of Signal Process. to Audio and Acoust. (WASPAA), Oct. 2021.

  10. Polynomial Matrix Eigenvalue Decomposition-Based Source Separation Using Informed Spherical Microphone Arrays

    Published in WASPAA, 2021

    This paper proposes a polynomial matrix eigenvalue decomposition-based source separation algorithm using informed spherical arrays.

    Recommended citation: V. W. Neo, C. Evers and P. A. Naylor, "Polynomial matrix eigenvalue decomposition-based source separation using informed spherical microphone arrays," in Proc. IEEE Workshop on Applications of Signal Process. to Audio and Acoust. (WASPAA), Oct. 2021.

  11. Polynomial Matrix Eigenvalue Decomposition of Spherical Harmonics for Speech Enhancement

    Published in ICASSP, 2021

    Suitable for spherical microphone arrays, this paper proposes a PEVD algorithm that uses only lower dimension eigenbeams for speech enhancement at a significantly lower cost while maintaining full performance of the original algorithm.

    Recommended citation: V. W. Neo, C. Evers, and P. A. Naylor, "Polynomial matrix eigenvalue decomposition of spherical harmonics for speech enhancement," in Proc. Intl. Conf. on Acoust., Speech and Signal Process. (ICASSP), Jun. 2021.

  12. Speech Dereverberation Performance of a Polynomial-EVD Subspace Approach

    Published in EUSIPCO, 2020

    This paper investigates the speech dereverberation performance of the PEVD-based speech enhancement algorithm, which was previously evaluated for noise reduction, speech intelligibility and quality performance.

    Recommended citation: V. W. Neo, C. Evers, and P. A. Naylor, "Speech dereverberation performance of a polynomial-EVD subspace approach," in Proc. Eur. Signal Process. Conf. (EUSIPCO), Aug. 2020.

  13. PEVD-based Speech Enhancement in Reverberant Environments

    Published in ICASSP, 2020

    This paper proposes a PEVD-based algorithm which is shown to be effective for speech enhancement in reverberant environments.

    Recommended citation: V. W. Neo, C. Evers and P. A. Naylor, "PEVD-based Speech Enhancement in Reverberant Environments," in Proc. IEEE Intl. Conf. on Acoust., Speech and Signal Process. (ICASSP), May 2020.

  14. Speech Enhancement Using Polynomial Eigenvalue Decomposition

    Published in WASPAA, 2019

    This paper proposes a speech enhancement algorithm using polynomial matrix eigenvalue decomposition in anechoic environments.

    Recommended citation: V. W. Neo, C. Evers and P. A. Naylor, "Speech enhancement using polynomial eigenvalue decomposition," in Proc. IEEE Workshop on Applications of Signal Process. to Audio and Acoust. (WASPAA), Oct. 2019.

  15. Second Order Sequential Best Rotation (SBR2) Algorithm With Householder Transformation for Polynomial Matrix Eigenvalue Decomposition (PEVD)

    Published in ICASSP, 2019

    This paper is about the use of Householder transformation on the SBR2 algorithm to speed up PEVD computation.

    Recommended citation: V. W. Neo and P. A. Naylor, "Second-order sequential best rotation algorithm with Householder transformation for polynomial matrix eigenvalue decomposition," in Proc. IEEE Intl. Conf. on Acoust., Speech and Signal Process. (ICASSP), May 2019.