Though this be hesitant, yet there is method in 't: Effects of disfluency patterns in neural speech synthesis for cultural heritage presentations

This study presents the results of two perception experiments aimed at evaluating the effect that specific patterns of disfluencies have on people listening to synthetic speech. We consider the particular case of Cultural Heritage presentations and propose a linguistic model to support the positioning of disfluencies throughout the utterances in the Italian language. A state-of-the-art speech synthesizer, based on Deep Neural Networks, is used to prepare a set of experimental stimuli and two different experiments are presented to provide both subjective evaluations and behavioural assessments from human subjects. Results show that synthetic utterances including disfluencies, predicted by a linguistic model, are identified as more natural and that the presence of disfluencies benefits the listeners' recall of the provided information.

Automated summary

This study presents the results of two perception experiments that evaluate the impact of specific patterns of disfluencies on listeners of synthetic speech. Focusing on Cultural Heritage presentations, the study proposes a linguistic model for positioning disfluencies in Italian language utterances. Utilizing a state-of-the-art speech synthesizer based on Deep Neural Networks, the study prepares experimental stimuli and conducts subjective evaluations and behavioral assessments. The results indicate that synthetic utterances with predicted disfluencies are perceived as more natural and improve the listeners' recall of the provided information.