Discovery Engine
A flexible scoring layer uses your listening signals and feedback to shape the queue.
- Balances similarity with freshness
- Responds to likes and skips
- Keeps variety across artists
- Adapts as your taste shifts
MusicWander reads your Last.fm signals and leans underground to surface artists you actually want to hear next.
A flexible scoring layer uses your listening signals and feedback to shape the queue.
See how your taste moves session to session, from mood and energy to era and depth.
Recommendations shift with your listening habits so each session feels right.
Begin with a single song and branch out through similarity, tags, and related artists.
Pushes lesser-known artists forward so the queue feels fresh, not chart-driven.
Keeps the queue fresh by down-weighting tracks you already know too well.
Exclude moods you are not feeling and steer the queue in real time.
Auto-refill and diversity controls keep things varied without losing the thread.
Your data never leaves your browser. No accounts, no tracking, no cloud storage.
Export your entire session to JSON. Import later to continue where you left off.
Parallel requests and smart caching keep discovery quick and responsive.
Install on any device. Works offline with cached data.
Start exploring underground music tailored to your taste.