Frontiers in cognition · 2026
Coordination dynamics in singing with human and artificial partners: the role of visual information.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Nishiyama Rina, Nonaka Tetsushi
Аннотация
INTRODUCTION: This study examined whether coordination dynamics in singing differ when a person sings along with a human partner vs. an artificial partner generated by voice synthesis software (VOCALOID 6). We further investigated how the presence or absence of visual information shapes these coordination patterns.
METHODS: Thirteen female participants sang a chorus excerpt of Silent Night in unison with a partner heard through earphones. The task crossed two partner types (a human singer and an artificial VOCALOID 6 singer) with two visual conditions (with and without partner video). Artificial-partner videos were created using an AI video platform (Hedra Character 3) that combined the artificial partner's synthesized voice with a still image of the human partner. Coordination was assessed through cross-correlation and Granger causality analyses of vocal amplitude envelopes, evaluated using linear mixed-effects models.
RESULTS: Singing with a human partner elicited stronger anticipatory synchronization and greater similarity in amplitude envelopes than singing with an artificial partner. With the human partner, participants tended to sing slightly ahead of their partner, indicative of anticipatory coordination. Notably, this tendency was further strengthened when visual information was available.
DISCUSSION: These findings indicate that the temporal variability of the human singing voice, together with the fine-grained bodily movements characteristic of naturalistic vocal production, provides perceptual information that supports anticipation and synchronization in joint singing.
METHODS: Thirteen female participants sang a chorus excerpt of Silent Night in unison with a partner heard through earphones. The task crossed two partner types (a human singer and an artificial VOCALOID 6 singer) with two visual conditions (with and without partner video). Artificial-partner videos were created using an AI video platform (Hedra Character 3) that combined the artificial partner's synthesized voice with a still image of the human partner. Coordination was assessed through cross-correlation and Granger causality analyses of vocal amplitude envelopes, evaluated using linear mixed-effects models.
RESULTS: Singing with a human partner elicited stronger anticipatory synchronization and greater similarity in amplitude envelopes than singing with an artificial partner. With the human partner, participants tended to sing slightly ahead of their partner, indicative of anticipatory coordination. Notably, this tendency was further strengthened when visual information was available.
DISCUSSION: These findings indicate that the temporal variability of the human singing voice, together with the fine-grained bodily movements characteristic of naturalistic vocal production, provides perceptual information that supports anticipation and synchronization in joint singing.