This is not a feature request, more a \"how would you do this?\" question......

So you fetch the lyrics
Take out words like \"the, like, a, my, our, us, we, you, they\"
Fetch Flickr images for 5-20% of remaining words
Run a visualizer over the images or integrate the images into visualizations
Enable the user to lock images to a song OR to have new images fetched each time.

Basically, you\'ve got a recipe for an instant music video. I want to make these, and I think that using a script and amaroK is probably the best way to do it. However, I don\'t know the first thing about making a script for amaroK. Could anyone give me a few pointers in the direction I would need to go in to make this work? I am thinking that there is great potential to recycle code here so that 100% new vis software does not need to be written. I imagine that the flickr lookup couldn\'t be that hard.