2020/06/08 by Hyeongju Kim, Hyeonseung Lee, Kim, Hyeongju +10 · 2 citations
Computer Science · Engineering · #Music and Audio Processing #Speech Recognition and Synthesis #Speech and Audio Processing #cs.CL #cs.LG #cs.SD #eess.AS
paper · pdf · doi:10.48550/arxiv.2006.04598
8 pages, 4 figures, Second workshop on Invertible Neural Networks, Normalizing Flows, and Explicit Likelihood Models (ICML 2020)
arxiv created 2020/07/02 · arxiv updated 2020/07/06
In recent years, various flow-based generative models have been proposed to generate high-fidelity waveforms in real-time. However, these models require either a well-trained teacher network or a number of flow steps making them memory-inefficient. In this paper, we propose a novel generative model called WaveNODE which exploits a continuous normalizing flow for speech synthesis. Unlike the conventional models, WaveNODE places no constraint on the function used for flow operation, thus allowing the usage of more flexible and complex functions. Moreover, WaveNODE can be optimized to maximize the likelihood without requiring any teacher network or auxiliary loss terms. We experimentally show that WaveNODE achieves comparable performance with fewer parameters compared to the conventional flow-based vocoders.