Unguided VR public-speaking training enhances your confidence - but does not improve your intonation
Ïo Valls-Ratés, Oliver Niebuhr, Pilar Prieto · Speech Prosody 2022 · 2022
Public speaking is essential in our daily life. However, standing in front of a crowd is more often than not challenging for people. VR simulations can help speakers meet this challenge. Our study employed a between-subjects design with a VR group (N=17) and a Non-VR group (N=14). Both groups gave a 2-minute speech in front of a live audience before (PRE) and after (POST) they practiced public speaking in front of a VR audience or alone in a classroom (Non-VR). Each group had three of these VR or Non-VR training sessions, one per week. Acoustic analyses of both groups’ PRE vs. POST prosodies show that 1) the two groups did not differ significantly in f0-related parameters as a function of training (f0 level, f0 range, f0 sd, f0 min/max), 2) the VR group has, unlike the Non-VR group, developed a stronger, clearer and more confident way of presenting, in terms of a longer talking time, fewer disfluent pauses, lower speaking rate, higher CPP and HNR, and lower jitter and shimmer levels. Thus, unguided VR training can help people give more persuasive speeches in real-life presentations, but we assume (consistent with previous research) that guided feedback is required to also improve people's speech melody.