4 papers
Continuous-Time Acoustic Modelling with Neural Controlled Differential Equations
Mattias Cross, Minghui Zhao, Anton Ragni
Text-to-speech (TTS) models commonly address text--speech alignment by expanding phone-level encoder states to frame-level decoder inputs using predicted durations. While this leng…
Discrete-Time Diffusion-Like Models for Speech Synthesis
Xiaozhou Tan, Minghui Zhao, Anton Ragni
Diffusion models have attracted a lot of attention in recent years. These models view speech generation as a continuous-time process. For efficient training, this process is typica…
Isometry Theorem for Continuous Quiver of Type
Xiaowen Gao, Minghui Zhao
The Isometry Theorem for continuous quiver of type plays an important role in persistent homology. In this paper, we shall generalize Isometry Theorem to continuous quiver of t…
Maximal Rigid Representations of Continuous Quivers of Type A with Automorphism
Xiaowen Gao, Minghui Zhao
Buan and Krause gave a classification of maximal rigid representations for cyclic quivers and counted the number of isomorphism classes. By using this result, we give a formula on…