2017 LatentSequenceDecompositions

• (Chan et al., 2017) ⇒ William Chan, Yu Zhang, Quoc V. Le, and Navdeep Jaitly. (2017). “Latent Sequence Decompositions.” In: Conference Track Proceedings of the 5th International Conference on Learning Representations (ICLR 2017).

Subject Headings: Latent Sequence Decompositions (LSD) System.

Notes

• OpenReview: https://openreview.net/forum?id=SyQq185lg

Cited By

• http://scholar.google.com/scholar?q=%222017%22+Latent+Sequence+Decompositions

Quotes

Abstract

We present the Latent Sequence Decompositions (LSD) framework. LSD decomposes sequences with variable lengthed output units as a function of both the input sequence and the output sequence. We present a training algorithm which samples valid extensions and an approximate decoding algorithm. We experiment with the Wall Street Journal speech recognition task. Our LSD model achieves 12.9% WER compared to a character baseline of 14.8% WER. When combined with a convolutional network on the encoder, we achieve 9.6% WER.

BibTeX

@inproceedings{2017_LatentSequenceDecompositions,
  author    = {William Chan and
               Yu Zhang and
               Quoc V. Le and
               Navdeep Jaitly},
  title     = {Latent Sequence Decompositions},
  booktitle = {Conference Track Proceedings of the 5th International Conference on Learning Representations
              (ICLR 2017)},
  publisher = {OpenReview.net},
  year      = {2017},
  url       = {https://openreview.net/forum?id=SyQq185lg},
}

---

	Author	volume	Date Value	title	type	journal	titleUrl	doi	note	year
2017 LatentSequenceDecompositions	Yu Zhang Quoc V. Le Navdeep Jaitly William Chan			Latent Sequence Decompositions						2017