A Parallelized Dynamic Programming Approach to Zero Resource Spoken Term Discovery

Bradley Oosterveld and Richard Veale and Matthias Scheutz

Describes a method for zero resource spoken term discovery, the process of deriving a set of linguistic tokens from a spoken corpus with no transcription. Our approach makes use of the Acoustic DP-ngram algorithm, and parallelization using GPUs.

@inproceedings{oosterveld2017icassp,
  title={A Parallelized Dynamic Programming Approach to Zero Resource Spoken Term Discovery},
  author={Bradley Oosterveld and Richard Veale and Matthias Scheutz},
  year={2017},
  booktitle={Proceedings of the 42nd IEEE International Conference on Acoustics, Speech, and Signal Processing},
  url={https://hrilab.tufts.edu/publications/oosterveld2017icassp.pdf}
}