iVector-based discriminative adaptation for automatic speech recognition

@article{Karafit2011iVectorbasedDA,
  title={iVector-based discriminative adaptation for automatic speech recognition},
  author={Martin Karafi{\'a}t and Luk{\'a}s Burget and Pavel Matejka and Ondrej Glembek and Jan {\vC}ernock{\'y}},
  journal={2011 IEEE Workshop on Automatic Speech Recognition & Understanding},
  year={2011},
  pages={152-157}
}
  • Martin Karafiát, Lukás Burget, +2 authors Jan Černocký
  • Published in
    IEEE Workshop on Automatic…
    2011
  • Computer Science
  • We presented a novel technique for discriminative feature-level adaptation of automatic speech recognition system. [...] Key Method To utilized iVectors for adaptation, Region Dependent Linear Transforms (RDLT) are discriminatively trained using MPE criterion on large amount of annotated data to extract the relevant information from iVectors and to compensate speech feature. The approach was tested on standard CTS data. We found it to be complementary to common adaptation techniques. On a well tuned RDLT system…Expand Abstract

    Create an AI-powered research feed to stay up to date with new papers like this posted to ArXiv

    Figures, Tables, and Topics from this paper.

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 23 CITATIONS

    Batch Normalization based Unsupervised Speaker Adaptation for Acoustic Models

    • Jiangyan Yi, Jian-hua Tao
    • Computer Science
    • 2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)
    • 2019
    VIEW 1 EXCERPT
    CITES METHODS

    Improving code-switching speech recognition with data augmentation and system combination

    VIEW 1 EXCERPT

    Speaker and Language Aware Training for End-to-End ASR

    SpeakerBeam: Speaker Aware Neural Network for Target Speaker Extraction in Speech Mixtures

    VIEW 1 EXCERPT

    Context Adaptive Neural Network Based Acoustic Models for Rapid Adaptation

    VIEW 2 EXCERPTS
    CITES METHODS & BACKGROUND

    Generalized Variability Model for Speaker Verification

    VIEW 1 EXCERPT

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 19 REFERENCES

    A compact model for speaker-adaptive training

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Constrained discriminative mapping transforms for unsupervised speaker adaptation

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL

    Unsupervised discriminative adaptation using discriminative mapping transforms

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL

    Minimum phoneme error based heteroscedastic linear discriminant analysis for speech recognition

    • Bing Zhang, Spyridon Matsoukas
    • Computer Science
    • Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005.
    • 2005
    VIEW 3 EXCERPTS
    HIGHLY INFLUENTIAL

    Full-covariance UBM and heavy-tailed PLDA in i-vector speaker verification

    VIEW 2 EXCERPTS

    Simplification and optimization of i-vector extraction

    VIEW 1 EXCERPT

    The 2007 AMI(DA) System for Meeting Transcription

    VIEW 1 EXCERPT