cs.SDNov 5, 2026

ProCraft: Instruction-Driven Music Production from Free-Form Text

Authors: Yuan-Chiao Cheng, Yi-Hsuan Yang, Vincent K. M. Cheung

Organizations: Georgia Institute of Technology · AI-CoRE, National Taiwan University · Sony Computer Science Laboratories, Tokyo

Abstract

Music production typically involves multiple editing operations that each require specialized tools and expertise. Existing text-guided music editing systems support few operations or instruments, operate on short clips, or handle only one operation per instruction, unlike real workflows where several changes are requested at once. Here, we present ProCraft, a lightweight 0.5B-parameter autoregressive codec language model that edits 30s music excerpts end-to-end from free-form instructions, including compound requests. ProCraft supports seven operations across eight instruments, from tasks with a unique target (e.g., track removal) to creative ones (e.g., accompaniment generation). We align intermediate representations with pretrained music embeddings and pitch and instrument targets to preserve content outside the edit, and gated cross-attention in every decoder layer to keep generation tied to the instruction. To train on compound requests, we introduce On-the-fly Editing-pair Synthesis, which generates paired instructions and audio combining track-level operations with effect changes. ProCraft performs competitively against task-specific systems on single-task editing, and outperforms four free-text editors on compound instructions for track removal, extraction and denoising on real (MoisesDB) and synthetic (Slakh) recordings. In a user study (16 listeners), ProCraft was rated highest on removal and denoising and on-par with a 16-times larger model on extraction and instrument transfer. Ablations further show that representation alignment and cross-attention reduce leakage from unwanted tracks and raise editing success rates, especially on out-of-domain data.

Figures & tables

Appendix figures & tables9 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. AURA: Unified Multimodal Framework for Conversational Music Editing

    Sep 13, 2026Quoc-Huy Trinh, Minh-Van Nguyen, Debesh JhaAudio EditingControllable Music Generation

  2. MusicFlux: Flexible Conditioning for Text-Guided Music Generation and Editing

    Nov 10, 2026Teysir Baoueb, Leonardo Boulitreau de Menezes Martins Marques, Paraskevas Stamatiadis +1Audio EditingControllable Music Generation

  3. FoleyForge: Free-form, Controllable, Rectified-flow Editing of Sound Effects

    Nov 5, 2026Giovana Morais, Hugo Flores García, Prem Seetharaman +4Audio EditingAudio-Language Models