2013 | OriginalPaper | Buchkapitel
Singing Like a Tenor without a Real Voice
verfasst von : Jochen Feitsch, Marco Strobel, Christian Geiger
Erschienen in: Advances in Computer Entertainment
Verlag: Springer International Publishing
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
We describe a multimedia installation that provides users with the experience to sing like a tenor from the early 20th century. The user defines vowels with her mouth but does not produce sound. The mouth shape is recognized and tracked by a depth-sensing camera and synthesized using a dedicated sound analysis using formants. Arm gestures are recognized and used to determine pitch and volume of an artificially generated voice. This synthesized voice is additionally modified by acoustic filters to sound like a singing voice from an old gramophone. The installation allows to scan the user’s face and to create an individual 3D model of a tenor character that is used to visualize the user performance.