Bi-directional Lattice Recurrent Neural Networks for Confidence Estimation

@article{Li2019BidirectionalLR,
  title={Bi-directional Lattice Recurrent Neural Networks for Confidence Estimation},
  author={Qiujia Li and Preben Ness and Anton Ragni and Mark John Francis Gales},
  journal={ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
  year={2019},
  pages={6755-6759}
}
  • Qiujia Li, Preben Ness, +1 author Mark John Francis Gales
  • Published in
    ICASSP - IEEE International…
    2019
  • Computer Science, Engineering
  • The standard approach to mitigate errors made by an automatic speech recognition system is to use confidence scores associated with each predicted word. In the simplest case, these scores are word posterior probabilities whilst more complex schemes utilise bi-directional recurrent neural network (BiRNN) models. A number of upstream and downstream applications, however, rely on confidence scores assigned not only to 1-best hypotheses but to all words found in confusion networks or lattices… CONTINUE READING
    12
    Twitter Mentions

    Citations

    Publications citing this paper.
    SHOWING 1-4 OF 4 CITATIONS

    C L ] 2 2 Ju l 2 01 9 ONMODELING ASR WORD CONFIDENCE

    VIEW 6 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    C L ] 5 O ct 2 01 9 ONMODELING ASR WORD CONFIDENCE

    VIEW 6 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    Uncertainty and confidence scores for sequence data

    VIEW 9 EXCERPTS
    CITES METHODS & BACKGROUND
    HIGHLY INFLUENCED

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 38 REFERENCES

    Confidence Estimation and Deletion Prediction Using Bidirectional Recurrent Neural Networks

    VIEW 4 EXCERPTS

    A post-processing system to yield reduced word error rates: Recognizer Output Voting Error Reduction (ROVER)

    • Jonathan G. Fiscus
    • Computer Science
    • 1997 IEEE Workshop on Automatic Speech Recognition and Understanding Proceedings
    • 1997
    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL

    Acoustic Modeling for Google Home

    VIEW 1 EXCERPT

    Attention is All you Need

    VIEW 1 EXCERPT

    Enriching Word Vectors with Subword Information

    VIEW 1 EXCERPT