Voices are arguably the most relevant sounds we hear in our everyday lives. Voices, in fact, are not only the carrier of speech as they convey a wealth of information about the speakers such as their identity, age and emotional state, just to name a few. Often, this vocal information can reinforce the cues we get from other sensory inputs and facilitate some processes such as emotion discrimination. In other cases, the features that we can extract from voices could be the only available cues if, for instance, visual information is not accessible or in the case of sensory deprivation. This dissertation aims at investigating voice perception in the adult human brain in the presence and absence of facial inputs. The results obtained from different experiments on: (i) voice perception in blind individuals, (ii) voice categorization investigated with EEG and (iii) multisensory face-voice integration in EEG will be presented and discussed in view of previous results.