EUVIMEDEUVIMED entdecken
Open Access · CC BY 3.0

Design and Application of an AIGC Music Generation Framework

Hongyu Zhao

International Journal of Knowledge Management · 20.02.2026 · 22 (1) · 1-18

DOI: 10.4018/ijkm.402193 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Knowledge Management, erschienen am 20.02.2026. Verfasst von Hongyu Zhao. Die Arbeit nennt 29 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 3.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In creative industries, AIGC music technologies hold promise for advancing organizational knowledge creation and reuse, but face barriers in resource efficiency, collaborative controllability, and alignment with knowledge management goals. This study develops a hybrid Transformer-GNN framework with a dual-loop control mechanism, tailored to support KM processes: knowledge capture, transfer, and reuse. Evaluations on MAESTRO-v3, Lakh-Clean, and a custom dataset show the framework outperforms mainstream models in Pitch Accuracy (85.3%), Harmonic Coherence (0.78), and Structure Completeness (0.73), with optimized resource use (36-hour training, 18.8 GB peak GPU memory). User studies confirm a 26% reduction in manual revisions, enhancing knowledge transfer between AI and human teams. This work provides a deployable AIGC solution bridging technical innovation and organizational KM needs in creative contexts.

Zitieren

APA

Zhao, H. (2026). Design and Application of an AIGC Music Generation Framework. International Journal of Knowledge Management, 22(1), 1-18. https://doi.org/10.4018/ijkm.402193

Vancouver

Zhao H. Design and Application of an AIGC Music Generation Framework. International Journal of Knowledge Management. 2026;22(1):1-18. doi:10.4018/ijkm.402193

BibTeX
@article{Zhao2026,
  author = {Zhao, Hongyu},
  title = {Design and Application of an AIGC Music Generation Framework},
  journal = {International Journal of Knowledge Management},
  year = {2026},
  volume = {22},
  number = {1},
  pages = {1--18},
  doi = {10.4018/ijkm.402193}
}
Kostenlos starten →