This chapter begins with a historical approach that explains why music and language have long been considered as modular and as such independent brain functions. It shows some methodological limitations of the neuropsychological and neuroimaging approach in interpreting differences between music and language neural substrates. The chapter advocates for the use of a truly comparative approach to study differences and commonalities between music and language. It then highlights several common functions and operations necessary in both language and music processing and presents the sharing resource hypothesis. The study of the effects of music training and music stimulation on language processing is a privileged avenue to unravel the relation of common operations and their specific nature at the algorithmic level. The chapter concludes with a focus on the temporal dimension and the dynamic nature of oscillatory activity and their role in temporal prediction in music and language.