2018 · 100 citations · 16 references
ComputingComputer ScienceComputational LinguisticsNatural Language ProcessingPhonologyLanguage ModelingLanguage ModelsN-gram AnalysisPattern RecognitionSpeech RecognitionLanguage Model RescoringN-gram RepresentationNeural Language ModelsRecurrent Neural Language ModelsRecurrent Neural Network Language ModelsSearch Space Pruning
Lattice-rescoring is a common approach to take advantage of recurrent neural language models in ASR, where a word-lattice is generated from 1st-pass decoding and the lattice is then rescored with a neural model, and an <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">n</i> -gram approximation method is usually adopted to limit the search space. In this work, we describe a pruned lattice-rescoring algorithm for ASR, improving the n-gram approximation method. The pruned algorithm further limits the search space and uses heuristic search to pick better histories when expanding the lattice. Experiments show that the proposed algorithm achieves better ASR accuracies while running much faster than the standard algorithm. In particular, it brings a 4x speedup for lattice-rescoring with 4-gram approximation while giving better recognition accuracies than the standard algorithm.
16
Kaldi Speech Recognition Toolkit
Daniel Povey · Infoscience (Ecole Polytechnique Fédérale de Lausanne) · 2024 · 4.9K citations · Full text
Computing, Linguistics, Acoustics +15
A study on data augmentation of reverberant speech for robust speech recognition
Tom Ko, Vijayaditya Peddinti, Daniel Povey et al. · 2017 · 917 citations
Purely Sequence-Trained Neural Networks for ASR Based on Lattice-Free MMI
Daniel Povey, Vijayaditya Peddinti, Daniel Gálvez et al. · 2016 · 795 citations