A Comparison of Linear and Nonlinear Dimensionality Reduction Methods Applied to Synthetic Speech

摘要

In this study a number of linear and nonlinear dimensionality reduction methods are applied to high dimensional represen-tations of synthetic speech to produce corresponding low di-mensional embeddings. Several important characteristics of the synthetic speech, such as formant frequencies and f0, are known and controllable prior to dimensionality reduction. The degree to which these characteristics are retained after dimensionality reduction is examined in visualisation and classification exper-iments. Results of these experiments indicate that each method is capable of discovering meaningful low dimensional represen-tations of synthetic speech and that the nonlinear methods may outperform linear methods in some cases.