A lot of what I've been thinking about in the contra dance music direction lately has been: as the rhythm player in a duo, how can I build the kind of full sound you generally need a larger group for? An idea in this direction that I haven't explored yet is instrumental vocals.

Most music people listen to has singing, but contra dance generally doesn't because it has calling instead. You can't have singing and calling at the same time without confusing the dancers, and the caller dropping out is unpredictable. There are some bands that have incorporated singing, though, generally as something they bring in near the end of a set. This can be really great when done well, and I've enjoyed dancing to singing by Gallimaufry (video), Chimney Swift (video), and Trip the Light Fantastic (video).

Another direction to take vocals in, though, would be something that had a more instrumental feel. If it doesn't sound like a voice, then I would expect you could do it at any time during a set, and not just after the caller drops out. Another thing I like about this idea, is that while singing requires building a set around a song and is very central and forward, something more instrumental-like could be a less-central addition.

(Why use voice? It's incredibly capable, flexible, and high bandwidth. And I'm already using my hands and feet for other things.)

The only thing I remember dancing to along these lines is Kate Barnes' high-energy vocal harmonizing on top of her piano playing at the end of a few sets; if anyone knows other examples in this direction I'd be very interested to hear about it.

Since the voice most naturally sounds like a voice, and conflicts with the caller, one option here would be to use some kind of effects. Since the vocals would not be central they could be pretty heavily modified and still provide an interesting addition. This would also offer a wider range of textures and atmospheric feelings than you can do with voice alone.

You could go even farther and use the voice as a MIDI controller. Extract the frequency to trigger notes, interpret the sound to implement multiple continuous controllers. Things that make me less excited about this direction include the latency of interpreting the frequency of notes as they're sung, not having the full acoustic freedom and control you normally have with the voice, and the sheer amount of work involved.

I'm not sure what sort of effects would be good to play with. Is this the sort of thing where people generally buy boxes, or wire the mic into a computer? I guess there's also both "what do you want to exploring and figuring out what effects you like" vs "what works well as part of a reliable stage setup". Advice welcome!