The correct answer is (b) Source extraction
For explanation I would say: Computation auditory scene analysis applies heuristic grouping rules developed using psychological observations to separate a source from a single channel of audio. It separates the sound into music, voice, noise, etc.