Pairwise Learning using Unsupervised Bottleneck Features for Zero-Resource Speech Challenge 2017 (System 1)

Dataset OPEN
Yuan, Yougen; Leung, Cheung-Chi; Xie, Lei; Chen, Hongjie; Ma, Bin; Li, Haizhou;
  • Publisher: Zenodo
  • Identifiers: doi: 10.5281/zenodo.809197
  • Subject: autoencoder | zenodo | pairwise learning | neural networks | zero-resource | Uncategorized | unsupervised bottleneck features
    acm: ComputingMethodologies_PATTERNRECOGNITION

<p>The system is for track1 alone.  We trained an antoencoder using unsupervised bottleneck features with word-pair information from Switchboard. The unsupervised bottleneck features was extracted from an extractor of multi-task learning deep neural networks (MTL-DNN). ... View more
Share - Bookmark