Learning Supervised Feature Transformations on Zero Resources for Improved Acoustic Unit Discovery

Michael HECK  Sakriani SAKTI  Satoshi NAKAMURA  

IEICE TRANSACTIONS on Information and Systems   Vol.E101-D   No.1   pp.205-214
Publication Date: 2018/01/01
Online ISSN: 1745-1361
DOI: 10.1587/transinf.2017EDP7175
Type of Manuscript: PAPER
Category: Speech and Hearing
acoustic unit discovery,  Bayesian nonparametrics,  feature transformation,  unsupervised subword modeling,  zero resource,  

Full Text: PDF(763.4KB)>>
Buy this Article

In this work we utilize feature transformations that are common in supervised learning without having prior supervision, with the goal to improve Dirichlet process Gaussian mixture model (DPGMM) based acoustic unit discovery. The motivation of using such transformations is to create feature vectors that are more suitable for clustering. The need of labels for these methods makes it difficult to use them in a zero resource setting. To overcome this issue we utilize a first iteration of DPGMM clustering to generate frame based class labels for the target data. The labels serve as basis for learning linear discriminant analysis (LDA), maximum likelihood linear transform (MLLT) and feature-space maximum likelihood linear regression (fMLLR) based feature transformations. The novelty of our approach is the way how we use a traditional acoustic model training pipeline for supervised learning to estimate feature transformations in a zero resource scenario. We show that the learned transformations greatly support the DPGMM sampler in finding better clusters, according to the performance of the DPGMM posteriorgrams on the ABX sound class discriminability task. We also introduce a method for combining posteriorgram outputs of multiple clusterings and demonstrate that such combinations can further improve sound class discriminability.