I know that we have a database interface for the images/videos of BANCA.
However, I think that the folder structure inside the audio dataset is not appropriate. I am not sure, why this is the case, and how the audio files are generated in the first place. As it seems (from the protocol files) there are only files for protocol "G", anyways.
I think, it would be a good idea to re-align the data to fit into the desired folder structure, and use bob.db.banca instead.
For timit, I am not sure if we have a database interface. If not, I would suggest to leave it inside bob.bio.spear -- except you think it might be useful outside the bob.bio packages. As far as I remember, the timit dataset was a subset of something, but I did not write the bob.bio.spear package.
I suggest to do the following:
If no one moves a finger to solve this one, we should delete these file lists.
Means that nobody use and cares about it.