Blog
About

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

      LSTM Language Models for LVCSR in First-Pass Decoding and Lattice-Rescoring

      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

          LSTM based language models are an important part of modern LVCSR systems as they significantly improve performance over traditional backoff language models. Incorporating them efficiently into decoding has been notoriously difficult. In this paper we present an approach based on a combination of one-pass decoding and lattice rescoring. We perform decoding with the LSTM-LM in the first pass but recombine hypothesis that share the last two words, afterwards we rescore the resulting lattice. We run our systems on GPGPU equipped machines and are able to produce competitive results on the Hub5'00 and Librispeech evaluation corpora with a runtime better than real-time. In addition we shortly investigate the possibility to carry out the full sum over all state-sequences belonging to a given word-hypothesis during decoding without recombination.

          Related collections

          Most cited references 10

          • Record: found
          • Abstract: not found
          • Conference Proceedings: not found

          Librispeech: An ASR corpus based on public domain audio books

            Bookmark
            • Record: found
            • Abstract: not found
            • Conference Proceedings: not found

            Strategies for training large scale neural network language models

              Bookmark
              • Record: found
              • Abstract: not found
              • Conference Proceedings: not found

              English Conversational Telephone Speech Recognition by Humans and Machines

                Bookmark

                Author and article information

                Journal
                01 July 2019
                Article
                1907.01030

                http://creativecommons.org/publicdomain/zero/1.0/

                Custom metadata
                eess.AS cs.LG cs.SD stat.ML

                Comments

                Comment on this article