Unsupervised Learning of Disentangled and Interpretable Representations from Sequential Data
This addresses the problem of improving speech processing tasks like speaker verification and recognition for applications in human-computer interaction, though it appears incremental as it builds on existing VAE methods.
The paper tackles unsupervised learning of disentangled and interpretable representations from sequential data, achieving a 35% reduction in word error rate for automatic speech recognition and outperforming an i-vector baseline for speaker verification.
We present a factorized hierarchical variational autoencoder, which learns disentangled and interpretable representations from sequential data without supervision. Specifically, we exploit the multi-scale nature of information in sequential data by formulating it explicitly within a factorized hierarchical graphical model that imposes sequence-dependent priors and sequence-independent priors to different sets of latent variables. The model is evaluated on two speech corpora to demonstrate, qualitatively, its ability to transform speakers or linguistic content by manipulating different sets of latent variables; and quantitatively, its ability to outperform an i-vector baseline for speaker verification and reduce the word error rate by as much as 35% in mismatched train/test scenarios for automatic speech recognition tasks.