Konferenzbeitrag
Evaluating Real-Time Pitch Estimation Algorithms for Creative Music Game Interaction
Vorschaubild nicht verfügbar
Volltext URI
Dokumententyp
Text/Conference Paper
Zusatzinformation
Datum
2023
Autor:innen
Zeitschriftentitel
ISSN der Zeitschrift
Bandtitel
Verlag
Gesellschaft für Informatik e.V.
Zusammenfassung
Music-based games are an important genre in the gaming community and have become increasingly popular with games like SingStar and Guitar Hero. These types of games are usually based on reactive game mechanics, where the player must hit a certain note at a certain time in order to score points. In this contribution, we present a game prototype that goes beyond purely music-reactive game mechanics and focuses more on the creative aspect of making music in games. In particular, we developed a jump-and-run game that can be controlled with a gaming controller but also uses the player’s singing voice to interact with the game world. To this end, we estimate the pitch of a microphone signal in real time and use it as a creative input to the game. This input can be used to control parts of the game world, for instance by singing and adding stair-like elements that allow the player to overcome obstacles and reach the end of a game level. With our game prototype, we demonstrate how game designers can incorporate musical challenges into a well-known game environment while motivating musicians to creatively explore and practice their musical skills. Furthermore, motivated by our game prototype, we evaluate different real-time pitch estimation algorithms using common MIR metrics on a publicly available dataset to analyse what works best for our gaming scenario.