The existence of a map** between emotions and speech prosody is commonly assumed.
We propose a Bayesian modelling framework to analyse this map**. Our models are fitted …

Speech is a natural interface for humans to interact with robots. Yet, aligning a robot's voice
to its appearance is challenging due to the rich vocabulary of both modalities. Previous …

Recent advances in multimodal training use textual descriptions to significantly enhance
machine understanding of images and videos. Yet, it remains unclear to what extent …

Novel text-to-speech systems can generate entirely new voices that were not seen during
training. However, it remains a difficult task to efficiently create personalized voices from a …

Neural Text-To-Speech (TTS) models achieve great performances regarding naturalness,
but modeling expressivity remains an ongoing challenge. Some success was found through …

Since neural Text-To-Speech models have achieved such high standards in terms of
naturalness, the main focus of the field has gradually shifted to gaining more control over the …

Modeling adequate features of speech prosody is one key factor to good performance in
affective speech classification. However, the distinction between the prosody that is induced …

Aesthetic appreciation is inherently multidimensional: many different stimulus dimensions
(eg, colors, shapes, sizes) contribute to our aesthetic experience. However, most studies in …

Conversational tones--the manners and attitudes in which speakers communicate--are
essential to effective communication. Amidst the increasing popularization of Large …

Modern TTS systems are capable of creating highly realistic and natural-sounding speech.
Despite these developments, the process of customizing TTS voices remains a complex …