Music is organised by socio-semiotic systems (music theory), represented by socio-semiotic systems (music notation), expanded by socio-semiotic systems (lyrics), and produced by socio-semiotic beings (humans), but music is not a socio-semiotic system. This is because, unlike genuine socio-semiotic systems, systems of content (meanings) cannot be systematically assigned to systems of expression (musical sounds).
In terms of Halliday's linear taxonomy of physical—biological—social—(socio-)semiotic systems, music is social, but not socio-semiotic. That is, music features value, but not symbolic value. According to Edelman's Theory of Neuronal Group Selection, perceptual categorisation is made on (+/–) values that have been adaptive to ancestors. On this basis, music can be understood as selecting values in the process of perception — values that weight all mental processes.
On the model of Halliday's Systemic Functional Linguistic Theory, then, one way to understand music is as a perceptual phenomenon that activates mental processes that range over emotions, desires and thoughts.