Talk by Prof. Juhan Nam (KAIST, South Korea)

[Translate to English:]

We are excited to announce that Professor Juhan Nam, director of the Music and Audio Computing Lab at Korea Advanced Institute of Science & Technology (KAIST, Daejeon, South Korea), will give a remote talk in our group on September 11^th at 10:00 am (CEST).

Abstract. Music Information Retrieval (MIR) is an interdisciplinary field that involves the computational analysis of music data to extract meaningful information, enabling applications such as music recommendation, transcription, classification, and musicological analysis. Recently, AI has been adopted as a new paradigm of computational methods in the field of MIR, achieving breakthroughs in many MIR tasks. This talk will present current research topics in the Music and Audio Computing Lab at KAIST, focusing on music-text association, music/audio generation, and human-AI ensemble.  In the topic of music-text association, we explore methods for describing music using tags or full sentences, and investigate joint audio-text embedding for text-based music retrieval. For music and audio generation, we will discuss two primary approaches: audio language models and diffusion models, which generate audio signals based on text prompts or other conditions. Finally, we will present our work on human-AI ensembles, which involves real-time, adaptive, and multimodal data processing to facilitate musical communication between humans and AI.

You can join us for the video-talk onsite at the Max Planck Institute for Empirical Aesthetics (conference room 416-419) or online here.

Looking forward to seeing you there!