Mustango project - congrats on winning the SAIL WAIC Award Top 30
We are pleased to share that Associate Professor Dorien Herremans, Assistant Professor Soujanya Poria, and their groups from ISTD have won the Super AI Leader (SAIL) Top 30 Award issued by the World Artificial Intelligence Conferences (WAIC), for their joint project ‘Mustango: controllable text-to-music’.
Background on the WAIC conference and WAIC SAIL Award
The World Artificial Intelligence Conference, or WAIC for short, is one of the most influential AI events within the global tech, science, and industry ecosystem. The event attracts the most prominent AI scientists, entrepreneurs, and government leaders globally. Notable attendees include Elon Musk (Tesla), Zhou Jingren (Alibaba), Raj Reddy (Turing Award winner), and Tom Mitchell (CMU).
The SAIL Award (Superior AI Leader Award) at WAIC is the highest honor of the conference and recognizes the most outstanding and breakthrough AI projects worldwide. The SAIL award embodies the concept of “pursuing excellence and leading the future.” Its selection and operation are guided by the principles of being “high-level, international, professional, market-oriented, and intelligent.”. This year's Super AI Leader Award, an honor for projects from all over the world with high recognition and reputation in the field of AI and with the significance of improving human welfare, has received over 200 submissions.
Mustango project
Mustango is a controllable text-to-music system that allows you to generate music audio by entering a text prompt. It is an open-source controllable text-to-music system. It allows the user to enter a detailed text prompt, including specific music theory instructions. For instance: “generate at song pop song with a moderato tempo and include chord sequences Am F C”. In terms of controllability, Mustango outperforms the leading systems MusicGen by Meta. The generated music is diversified and encompasses many different genres, from western pop to Chinese music, Indian, and opera, just to name a few. Mustango arose from a collaboration between Prof. Dorien Herremans and Prof. Soujanya Poria at SUTD. A demo is available here: https://huggingface.co/spaces/declare-lab/mustango
Paper: Melechovsky, J., Guo, Z., Ghosal, D., Majumder, N., Herremans, D., & Poria, S. (2024). Mustango: Toward controllable text-to-music generation. Proc. Of the Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL). Preprint link: https://arxiv.org/abs/2311.08355