From the first days of life, babies appear to be naturally attracted to human faces and voices. These early biases maximize their social interaction and experience of social stimuli, leading to an impressive neurocognitive development of social perception in their first year. Recent advances in neuroimaging methods have revealed patterns of voice sensitivity in the infant’s temporal cortex. These patterns resemble the ones observed in adults, and their voice sensitivity increases rapidly in the first few months of life. Voice sensitivity in the infant temporal cortex is observed cross-culturally but seems to be altered in infants at risk of developing autism spectrum disorder. Activation of the infant social brain is also modulated by the emotional content of human vocalizations in regions involved in emotion processing in adulthood, such as the temporal voice-sensitive area, amygdala, and orbitofrontal cortex. Very young infants also appear to be sensitive to the integration of facial and vocal cues of gender, age, identity, emotions, and speech articulation. These skills are present from the first months of life and are increasingly more sophisticated as infants get closer to their first birthday. This chapter discusses the most recent findings in the early development of voice processing in the infant brain, as well as the emergence of audiovisual integration in social perception.