2017 · 11 citations · 11 references
ComputingComputer ScienceComputational LinguisticsMachine LearningNatural Language ProcessingDeep LearningLanguage ModelingLanguage ModelsN-gram AnalysisSpeech RecognitionDirichlet Process ModelsNormalized Mutual InformationPhoneme RecognitionPhonotactic ProbabilityPitman-yor ProcessSegmenting Spoken Words Into Phonemes
Recent work on Acoustic Unit Discovery (AUD) has led to the development of a non-parametric Bayesian phone-loop model where the prior over the probability of the phone-like units is assumed to be sampled from a Dirichlet Process (DP). In this work, we propose to improve this model by incorporating a Hierarchical Pitman-Yor based bigram Language Model on top of the units' transitions. This new model makes use of the phonotactic context information but assumes a fixed number of units. To remedy this limitation we first train a DP phone-loop model to infer the number of units, then, the bigram phone-loop is initialized from the DP phone-loop and trained until convergence of its parameters. Results show an absolute improvement of 1–2%on the Normalized Mutual Information (NMI) metric. Furthermore, we show that, combined with Multilingual Bottleneck (MBN) features the model yields a same or higher NMI as an English phone recogniser trained on TIMIT.
11
Kaldi Speech Recognition Toolkit
Daniel Povey · Infoscience (Ecole Polytechnique Fédérale de Lausanne) · 2024 · 4.9K citations · Full text
Computing, Linguistics, Acoustics +15
Variational inference for Dirichlet process mixtures
David M. Blei, Michael I. Jordan · Bayesian Analysis · 2006 · 1.5K citations · Full text
The Infinite Gaussian Mixture Model
Carl Edward Rasmussen · 1999 · 992 citations
SJ Young, G Evermann, Mark Gales et al. · Cambridge University Engineering Department Publications Database · 2006 · 626 citations