This is a tool I wrote while working on the project PolFon (see polfon.upol.cz) to help me manage a large database of Polish speech sound samples.

It is written in the Praat scripting language and contains a lot of hacks and examples of bad coding, however it has been good enough for my purposes. In the future I might consider rewriting the code to be more sane and also provide a proper manual so that it would be useful for others but at the moment I don't have the time to do that.

You can try running the script in Praat and see what happens. The corpus builder can be operated with the mouse or with the keyboard (in that case I attempted to use some vim-like bindings: hjkl + Alt, Shift, Ctlr). At the moment, you have to go to the code itself to find the keybindings, or just play with the app with some data you are not afraid to lose. In the Extracts screen you can bring up a help a list of shortcuts by pressing "?" (in other words "Shift-/") on the keyboard.