The Experts below are selected from a list of 303 Experts worldwide ranked by ideXlab platform

J. Sorensen - One of the best experts on this subject based on the ideXlab platform.

  • Hierarchical pattern classification for high performance text-independent speaker verification systems
    Proceedings of ICASSP '94. IEEE International Conference on Acoustics Speech and Signal Processing, 1994
    Co-Authors: J. Sorensen, M. Savic
    Abstract:

    A new algorithm, the hierarchical speaker verification algorithm, is introduced. This algorithm employs a set of unique mapping functions determined from an enrolment utterance that characterize the target voice as a multidimensional Martingale random walk process. For sufficiently long verification utterances, the central limit theorem insures that the accumulated scores for the target speaker will be distributed normally about the origin. Impostor speakers, which violate the Martingale Property, are distributed arbitrarily and widely scattered in the verification space. Excerpts of verification performance experiments are given and extensions to the algorithm for handling noisy channels and speaker template aging are discussed.

  • Hierarchical pattern classification for high performance text- independent speaker verification systems
    ICASSP IEEE International Conference on Acoustics Speech and Signal Processing - Proceedings, 1994
    Co-Authors: J. Sorensen, Milan Savić
    Abstract:

    A new algorithm, the hierarchical speaker verification \nalgorithm, is introduced. This algorithm employs a set of \nunique mapping functions determined from an enrollment \nutterance that characterize the target voice as a \nmultidimensional Martingale random walk process. For \nsufficiently long verification utterances, the central limit \ntheorem insures that the accumulated scores for the target \nspeaker will be distributed normally about the origin. Impostor\nspeakers, which violate the Martingale Property, are \ndistributed arbitrarily and widely scattered in the \nverification space. Excerpts of verification performance \nexperiments are given and extensions to the algorithm for \nhandling noisy channels and speaker template aging are \ndiscussed. (Author abstract) 14 Refs.

M. Savic - One of the best experts on this subject based on the ideXlab platform.

  • Hierarchical pattern classification for high performance text-independent speaker verification systems
    Proceedings of ICASSP '94. IEEE International Conference on Acoustics Speech and Signal Processing, 1994
    Co-Authors: J. Sorensen, M. Savic
    Abstract:

    A new algorithm, the hierarchical speaker verification algorithm, is introduced. This algorithm employs a set of unique mapping functions determined from an enrolment utterance that characterize the target voice as a multidimensional Martingale random walk process. For sufficiently long verification utterances, the central limit theorem insures that the accumulated scores for the target speaker will be distributed normally about the origin. Impostor speakers, which violate the Martingale Property, are distributed arbitrarily and widely scattered in the verification space. Excerpts of verification performance experiments are given and extensions to the algorithm for handling noisy channels and speaker template aging are discussed.

Milan Savić - One of the best experts on this subject based on the ideXlab platform.

  • Hierarchical pattern classification for high performance text- independent speaker verification systems
    ICASSP IEEE International Conference on Acoustics Speech and Signal Processing - Proceedings, 1994
    Co-Authors: J. Sorensen, Milan Savić
    Abstract:

    A new algorithm, the hierarchical speaker verification \nalgorithm, is introduced. This algorithm employs a set of \nunique mapping functions determined from an enrollment \nutterance that characterize the target voice as a \nmultidimensional Martingale random walk process. For \nsufficiently long verification utterances, the central limit \ntheorem insures that the accumulated scores for the target \nspeaker will be distributed normally about the origin. Impostor\nspeakers, which violate the Martingale Property, are \ndistributed arbitrarily and widely scattered in the \nverification space. Excerpts of verification performance \nexperiments are given and extensions to the algorithm for \nhandling noisy channels and speaker template aging are \ndiscussed. (Author abstract) 14 Refs.

David Criens - One of the best experts on this subject based on the ideXlab platform.

C C Heyde - One of the best experts on this subject based on the ideXlab platform.