Distributed acoustic modeling with back-off n-grams

@article{Chelba2012DistributedAM,
  title={Distributed acoustic modeling with back-off n-grams},
  author={Ciprian Chelba and Peng Xu and Fernando C Pereira and Thomas J. Richardson},
  journal={2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
  year={2012},
  pages={4129-4132}
}
  • Ciprian Chelba, Peng Xu, +1 author Thomas J. Richardson
  • Published in
    IEEE International Conference…
    2012
  • Computer Science
  • The paper proposes an approach to acoustic modeling that borrows from n-gram language modeling in an attempt to scale up both the amount of training data and model size (as measured by the number of parameters in the model) to approximately 100 times larger than current sizes used in ASR. Dealing with unseen phonetic contexts is accomplished using the familiar back-off technique used in language modeling due to implementation simplicity. The new acoustic model is estimated and stored using the… CONTINUE READING

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

    Citations

    Publications citing this paper.

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 18 REFERENCES

    Query language modeling for voice search

    VIEW 5 EXCERPTS

    MapReduce: simplified data processing on large clusters

    VIEW 11 EXCERPTS
    HIGHLY INFLUENTIAL

    Distributed Language Models

    VIEW 1 EXCERPT

    The Unreasonable Effectiveness of Data

    VIEW 1 EXCERPT

    Boosted MMI for model and feature-space discriminative training

    VIEW 1 EXCERPT

    Progress in the CU-HTK broadcast news transcription system

    VIEW 2 EXCERPTS

    Confidence measures for speech recognition: A survey

    VIEW 1 EXCERPT