56
views
0
recommends
+1 Recommend
0 collections
    0
    shares
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      Learning Hidden Unit Contributions for Unsupervised Acoustic Model Adaptation

      Preprint

      Read this article at

      Bookmark
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Abstract

          This work presents a broad study on the adaptation of neural network acoustic models by means of learning hidden unit contributions (LHUC) -- a method that linearly re-combines hidden units in a speaker- or environment-dependent manner using small amounts of unsupervised adaptation data. We also extend LHUC to a speaker adaptive training (SAT) framework that leads to more adaptable DNN acoustic model, which can work in both a speaker-dependent and a speaker-independent manner, without the requirement to maintain auxiliary speaker-dependent feature extractors or to introduce significant speaker-dependent changes to the DNN structure. Through a series of experiments on four different speech recognition benchmarks (TED talks, Switchboard, AMI meetings and Aurora4) and over 270 test speakers we show that LHUC in both its test-only and SAT variants results in consistent word error rate reductions ranging from 5% to 23% relative depending on the task and the degree of mismatch between training and test data. In addition we have investigated the effect of the amount of adaptation data per speaker, the quality of adaptation targets when estimating transforms in an unsupervised manner, the complementarity to other adaptation techniques, one-shot adaptation, and an extension to adapting DNNs trained in a sequence discriminative manner.

          Related collections

          Author and article information

          Journal
          1601.02828

          Theoretical computer science,Artificial intelligence,Graphics & Multimedia design

          Comments

          Comment on this article