▎ Sonosketch turns hummed, whistled, or sung melodies into instrument performances. Sketch an
▎ idea with your voice, specify a target instrument (piano, guitar, sax, ...), and get audio
▎ that follows your melody's contour and rhythm. Built on ACE-Step 1.5 XL-SFT cover mode with
▎ tuned defaults, seed control, and per-output run logs for reproducibility
Research covers technology stack (ACE-Step vs MusicGen), feature landscape, MusicGen melody architecture analysis, and domain pitfalls. Key finding: ACE-Step 1.5 cover mode is the proven primary engine; MusicGen via HF transformers has fundamental conditioning flaws. |
||
|---|---|---|
| .planning | ||