Mostly we use keyboard, mouse, touch screen, etc. to control computers. Why not use singing? :)
Mice, joysticks and similar devices provide a position in one or more coordinates. Singing voice is a natual human expression and can be treated as a time-frequency energy distribution. We can extract multiple features such as fundamental frequency, intensity, harmonic profile, etc. and use them as input coordinates for computers. And it can be fun.
The research questions are:
- would it be practical at all?
- what would be the most usable features?
- what kind of application can be controlled by singing?
- could also discrete time events be utilizes?
- could be generalize singing also to rythm pattern, such as beatbox?
Ideas of application:
- music sketching
- sing a melody or beatbox a rhythm and convert it into a pattern for further usage
- games
- like ping pong
- pitch would control the position of the racket