ReCAB-VAE: Gumbel-Softmax Variational Inference Based on Analytic Divergence

no code implementations • 9 May 2022 • Sangshin Oh, Seyun Um, Hong-Goo Kang

In this work, we present a relaxed categorical analytic bound (ReCAB), a novel divergence-like metric which corresponds to the upper bound of the Kullback-Leibler divergence (KLD) of a relaxed categorical distribution.

Speech Synthesis Text-To-Speech Synthesis +1