It is meant to make the programmer's sound playback tasks simpler.
The programmer gives SDL_sound a filename, or feeds it data directly from one of many sources, and then reads the decoded waveform data back at her leisure.
If resource constraints are a concern, SDL_sound can process sound data in programmer-specified blocks.
Alternately, SDL_sound can decode a whole sound file and hand back a single pointer to the whole waveform.
SDL_sound can also handle sample rate, audio format, and channel conversion on-the-fly and behind-the-scenes, if the programmer desires.