Analysis and Assessment of Controllability of an Expressive Deep Learning-Based TTS System

Informatics

Noé Tits¹  Thierry Dutoit²  Kevin El Haddad² 

[1] Flowchase SRL, 1348 Ottignies-Louvain-la-Neuve, Belgium;TCTS Lab, University of Mons, Place du Parc 20, 7000 Mons, Belgium;

关键词: deep learning;    speech synthesis;    style interpolation;    perception;    artificial intelligence: affective computing;    emotion;   

DOI  :  10.3390/informatics8040084

来源: DOAJ

【 摘 要 】

In this paper, we study the controllability of an Expressive TTS system trained on a dataset for a continuous control. The dataset is the Blizzard 2013 dataset based on audiobooks read by a female speaker containing a great variability in styles and expressiveness. Controllability is evaluated with both an objective and a subjective experiment. The objective assessment is based on a measure of correlation between acoustic features and the dimensions of the latent space representing expressiveness. The subjective assessment is based on a perceptual experiment in which users are shown an interface for Controllable Expressive TTS and asked to retrieve a synthetic utterance whose expressiveness subjectively corresponds to that a reference utterance.

【 授权许可】

Unknown