Music Therapists

Right now, music therapy is still mostly human-led. Some new AI tools are being tested, but none have replaced a therapist. For example, researchers have built “AI music therapy” systems that track a person’s mood (using heart rate or facial cues) and then pick or tweak songs to help calm them ^{[1] [1]}.

But these are mostly lab demos, not everyday tools. In general healthcare, AI note-taking tools are starting to appear: one study found that AI could write short visit summaries and help doctors save time ^[1]. In the future, similar tools might help music therapists document sessions faster.

Wearable sensors plus AI can also monitor stress or emotions (e.g. using heart rate or skin conductance) ^[1]. However, such continuous monitoring raises privacy and security questions ^[1], and patients must consent. Crucially, tasks like improvising music or tuning therapy to a client’s unique needs still need human skill.

As ONET notes, music therapists must “compose, arrange, or adapt music for music therapy” and “identify and respond to emergency physical or mental health situations”* ^{[2] [2]}. These creative, judgment-based tasks have seen almost no automation. In short, most support is still a human making decisions – AI might help with things like song selection or notes, but it doesn’t replace the caring, creative person at the piano.

AI tools for therapy are still new and not widely sold to music therapists. Cutting-edge systems (like adaptive playlists or smart sensors) are mostly in research labs, not in clinics. Building a custom AI system is expensive, and most music therapy centers are small.

Because music therapists earn a modest wage (around $65K/year ^[2]) and jobs are growing, clinics often invest in more therapists rather than tech. (BLS projects music therapy jobs to grow “much faster than average” ^[2].) This means there’s less pressure to cut staff with machines.