Please ignore everything in this missive, including this sentence.
Simon,
IMHO, value is entirely subjective. Comparisons are an important part of value judgment. If I got what I considered to be $50 per year of value out of this project, I still wouldn't pay that much if I could get it for less. It's easily worth $2 per month to me, perhaps considerably more, but dmclellan01 has a point - someone else could make a version for less. And, if the audience is big enough, they probably will. But I don't think your low subscription rate is a failure in marketing, I think your audience is actually very small, as most people prefer just to listen to music rather than tinker with settings. And many (most?) of those who like to tinker probably don't land on Google Music as their music player of choice, since there are so many roll-your-own options that fit the tinkering crowd better. So your audience is limited to those who want to tinker with their music, but don't want to spend too much effort on it (maybe add those who want cloud support for their tinkering but have limited bandwidth where they might keep a server).
So if all that is close to correct, and you're not bringing in enough revenue, the correct response might be to RAISE the price, since lowering it will not likely extend your reach enough to make up for the lowered per-user rate.
On the other hand, I'd personally rather you not take that route.
If you do, and it works, maybe you could remember those that encouraged you with a grandfathered rate or something. And if you do it and it doesn't work, forget I said anything! ;-)
- Trevor.