The purpose of this section is to give some hints about sound management. See also our section Audio with SDL and Helix-OSDL (audio for the Nintendo DS).
As a rule of thumb, always keep your sounds stored with the best quality (using raw samples i.e. non encoding and high sampling frequencies), even if they take a lot more space, and transform them while they are still at their best quality. Lossy encoding and downsampling should be performed only at the very end of your sound pipeline.
Under GNU/Linux, for casual needs one may use krecord (the KDE sound recorder) which we found more useful than its Gnome counterpart. It produces WAVE files. wavrec does not seem to generate valid WAVE files.
When recording for application playback (e: for a game), we advise to use a not-too-bad microphone (ex: we are using a Shure JSH W8900D) and adequate software, like Audacity.
Of course, using a hi-fi allows to better ensure that your sound quality is sufficient. Headphones help checking the stereo and panning settings.
Lighter than xmms, wavplay or playwave (test program from SDL_mixer) can be used.
ogg123 does it for you.
oggenc welcome-to-OSDL.wav produces welcome-to-OSDL.ogg.
LAME is a powerful encoder.
Apart Audacity, one may use the sox tool, which is very rich and command-line based, thus very useful for scripted transformations (ex: massive conversions automated by makefiles).
If you have information more detailed or more recent than those presented in this document, if you noticed errors, neglects or points insufficiently discussed, drop us a line!