Hello Everyone,

We wish to undertake a new project for college (we are students of Computer Science at University of Buenos Aires) and the idea is to develop a virtual sound card driver which is capable of capturing the sound output generated by any standard Linux application, in order to do something with these data (for example dumping them to a file).

We are not sure about which level this driver should be implemented at. We would like to avoid, when possible, any low level or hardware related dealing.  One approach that we've been considering is to write something at the same level of ALSA or OSS (i.e. with the same API), which somehow could be registered in the system and made available to the "client" app's  as another selectable sound system.

We are a little confused about the Linux's sound architecture and would like to have some starting point to begin dealing with this project.  We would appreciate any help.

Thanks in advance.