Categories
Pages
-

DBIS

MuCo – Music Composer: Can AI Rival Human Creativity?

March 28th, 2023

Thesis Type
  • Master
Status
Open
Presentation room
Seminar room I5 6202
Supervisor(s)
Stefan Decker
Advisor(s)
Yongli Mou
Contact
mou@dbis.rwth-aachen.de

With its advanced capabilities in Natural Language Processing (NLP), ChatGPT is now sweeping the world and attracting significant attention and interest from researchers and developers worldwide. Meanwhile, researchers have started exploring the potential of AI in other creative fields, such as music composition. The use of generative models for music composition has also been gaining popularity, with various models being developed to automate the composition process and enhance creativity.

The main objectives of this thesis are to:

  • Explore the current state-of-the-art generative models for music composition,
  • Investigate the potential of Large Language Models (LLMs) in generative models for music composition, including lyrics generation, music comments generation, and direct music generation.

If you are interested in this thesis, do not hesitate to contact us via mou@dbis.rwth-aachen.de

 

Seed literature and links
[1] LLaMA: Open and Efficient Foundation Language Models. Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, Aurelien Rodriguez, Armand Joulin, Edouard Grave, Guillaume Lample. https://arxiv.org/abs/2302.13971v1
[2] Self-Instruct: Aligning Language Model with Self Generated Instructions. Yizhong Wang, Yeganeh Kordi, Swaroop Mishra, Alisa Liu, Noah A. Smith, Daniel Khashabi, Hannaneh Hajishirzi. https://arxiv.org/abs/2212.10560
[4] https://github.com/tatsu-lab/stanford_alpaca
[5] https://github.com/tloen/alpaca-lora
[6] Schneider F, Jin Z, Schölkopf B. Mo\^ usai: Text-to-Music Generation with Long-Context Latent Diffusion. arXiv preprint arXiv:2301.11757. 2023 Jan 27.
[7] Agostinelli A, Denk TI, Borsos Z, Engel J, Verzetti M, Caillon A, Huang Q, Jansen A, Roberts A, Tagliasacchi M, Sharifi M. Musiclm: Generating music from text. arXiv preprint arXiv:2301.11325. 2023 Jan 26.
[8] Mittal G, Engel J, Hawthorne C, Simon I. Symbolic music generation with diffusion models. arXiv preprint arXiv:2103.16091. 2021 Mar 30.
[9] https://github.com/lucylow/Stochastic_SoundCloud
[10] https://github.com/Natooz/MidiTok


Prerequisites:

Deep Knowledge of Deep Learning
Programming language – Python
Deep Learning Framework – PyTorch