Learning Supervised Feature Transformations on Zero Resources for Improved Acoustic Unit Discovery
-
- HECK Michael
- Augmented Human Communication Laboratory, Graduate School of Information Science, Nara Institute of Science and Technology
-
- SAKTI Sakriani
- Augmented Human Communication Laboratory, Graduate School of Information Science, Nara Institute of Science and Technology
-
- NAKAMURA Satoshi
- Augmented Human Communication Laboratory, Graduate School of Information Science, Nara Institute of Science and Technology
Abstract
<p>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.</p>
Journal
-
- IEICE Transactions on Information and Systems
-
IEICE Transactions on Information and Systems E101.D (1), 205-214, 2018
The Institute of Electronics, Information and Communication Engineers
- Tweet
Keywords
Details 詳細情報について
-
- CRID
- 1390001204380977920
-
- NII Article ID
- 130006301188
-
- ISSN
- 17451361
- 09168532
-
- Text Lang
- en
-
- Data Source
-
- JaLC
- Crossref
- CiNii Articles
- KAKEN
-
- Abstract License Flag
- Disallowed