Text-to-Music
Generating complete music — melody, instruments, vocals, structure — from a text description. Models like Suno turn 'dreamy synthwave with female vocals about city nights, 100 BPM' into a fully produced song in minutes, including lyrics if you want. Prompts work best describing genre, mood, instrumentation, tempo, and vocal style rather than abstract feelings. Creators use it for video soundtracks, podcast intros, brand jingles, and even full releases. Example workflow on VAR2: describe the vibe, generate a few candidates, then extend or remix the best one until it fits.