A new efficient measure for accuracy prediction and its application to multistream-based unsupervised adaptation

Tetsuji Ogawa, Sri Harish Mallidi, Emmanuel Dupoux, Jordan Cohen, Naomi H. Feldman, Hynek Hermansky

    研究成果: Conference contribution

    1 引用 (Scopus)

    抜粋

    A new efficient measure for predicting estimation accuracy is proposed and successfully applied to multistream-based unsupervised adaptation of ASR systems to address data uncertainty when the ground-truth is unknown. The proposed measure is an extension of the M-measure, which predicts confidence in the output of a probability estimator by measuring the divergences of probability estimates spaced at specific time intervals. In this study, the M-measure was extended by considering the latent phoneme information, resulting in an improved reliability. Experimental comparisons carried out in a multistream-based ASR paradigm demonstrated that the extended M-measure yields a significant improvement over the original M-measure, especially under narrow-band noise conditions.

    元の言語English
    ホスト出版物のタイトル2016 23rd International Conference on Pattern Recognition, ICPR 2016
    出版者Institute of Electrical and Electronics Engineers Inc.
    ページ2222-2227
    ページ数6
    ISBN(電子版)9781509048472
    DOI
    出版物ステータスPublished - 2017 4 13
    イベント23rd International Conference on Pattern Recognition, ICPR 2016 - Cancun, Mexico
    継続期間: 2016 12 42016 12 8

    Other

    Other23rd International Conference on Pattern Recognition, ICPR 2016
    Mexico
    Cancun
    期間16/12/416/12/8

      フィンガープリント

    ASJC Scopus subject areas

    • Computer Vision and Pattern Recognition

    これを引用

    Ogawa, T., Mallidi, S. H., Dupoux, E., Cohen, J., Feldman, N. H., & Hermansky, H. (2017). A new efficient measure for accuracy prediction and its application to multistream-based unsupervised adaptation. : 2016 23rd International Conference on Pattern Recognition, ICPR 2016 (pp. 2222-2227). [7899966] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICPR.2016.7899966