AES Show 2024 NY

There is a renewed need to study the development of AI systems that the artistic community can embrace as empowering tools rather than replacements for human creativity. Previous research suggests that novel AI technologies can greatly help artists expand the range of their musical expressivity and display new kinds of virtuosity during live music performances[1,2]. However, the advent of powerful AI systems, in particular generative models, has also attracted criticism and sparked concern among artists who fear for their artistic integrity and financial viability[3,4]. This is further exacerbated by a widening gap in technological innovation between private companies and research-focused academic institutions.

In this context, we need to pay specific attention to topics such as artistic consent, data collection[5], as well as audience approval. Furthermore, we deeply believe in the importance of integrating lighting and visual arts to effectively convey the role and impact of AI-generated content in human-AI performances to broader audiences.

This workshop will bring together researchers and professionals from music, lighting, visual arts, and artificial intelligence to explore the latest advancements in AI technologies and their transformative potential for live music performances. In particular, discussions will touch on the controllability requirements of AI-augmented instruments, the associated visualization methods, and the sociocultural impact of these technologies[6]. We will focus on the limitations of such technologies, as well as ethical considerations. As an example, we will also discuss the outcomes of the innovative Human-AI co-created concert that we produced with Jordan Rudess on September, 21st, 2024.

References:
[1] Blanchard, Lancelot, Naseck, Perry, Egozy, Eran, and Paradiso, Joe. “Developing Symbiotic Virtuosity: AI-augmented Musical Instruments and Their Use in Live Music Performances.” MIT Press, 2024.
[2] Martelloni, Andrea, McPherson, Andrew P, and Barthet, Mathieu. “Real-time Percussive Technique Recognition and Embedding Learning for the Acoustic Guitar.” arXiv, 2023.
[3] Morreale, Fabio. “Where does the buck stop? Ethical and political issues with AI in music creation.” Transactions of the International Society for Music Information Retrieval, 2021.
[4] Rys, Dan. “Billie Eilish, Pearl Jam, Nicki Minaj Among 200 Artists Calling for Responsible AI Music Practices.” Billboard, April 2, 2024.
[5] Morreale, Fabio, Sharma, Megha, and Wei, I-Chieh. “Data Collection in Music Generation Training Sets: A Critical Analysis.” International Society for Music Information Retrieval, 2023.
[6] Born, Georgina, Morris, Jeremy, Diaz, Fernando, and Anderson, Ashton. “Artificial intelligence, music recommendation, and the curation of culture". White paper: University of Toronto, Schwartz Reisman Institute for Technology and Society, CIFAR AI & Society program, 2021.

Come join the team from the Tony award winning stage play Stereophonic at AES NY 2024, to learn what it takes to put a 1970s recording studio on stage!

One of the considerations in electronic instrument design has always been expressive control, how to give the player a deeper more intuitive connection with an instrument's sound engine. MIDI Polyphonic Expression (MPE) was adopted by the MIDI Manufacturers Association in 2018 as an enhancement to the original specification. Since then manufacturer support has grown and several new instruments incorporate this type of gestural control. This panel will examine the creative possibilities of MPE from a sound design persecutive, exploring strategies for three dimensional control of sound parameters, their practical ranges and opportunities for reinventing existing instrument categories with continuous pitch control as well as more complex timbral effects.

As applications of artificial intelligence and machine learning become prevalent across the music technology industry, this panel will examine the ways that these technologies are influencing the design of new electronic instruments. As part of the discussion we’ll looking at current instruments as well and the potential for new designs.

Reports from the DEI committee and subcommittees will be given. The floor will then be opening for questions and discussion.

Metadata is essential in audio files for archiving, discovering, and accessing content. But in current multiplatform environments, metadata management can become difficult and complex. One solution is to implement a monolithic, centralised system, but this is often impractical and overly rigid in dynamic production workflows, since different clients or stakeholders may prefer specific delivery systems. As a result, media managers often have to deal with disparate distribution systems such as file-sharing services, e-mail attachments, download URLs, proprietary APIs, etc. The result of many of these distribution processes is that the audio content and its descriptive metadata often become separated, making the content potentially far less usable.

Embedding metadata in your audio files allows for easier, "one prong" delivery of content, in which you can obtain more robustness as the audio travels downstream along your workflows and into your client's systems. Furthermore, embedding metadata can also simplify system integration from a two-way process to a one-way process, where the receiving system does not need to be aware of specific requirements from the transmitting system.

In this workshop we will explore how the New York Public Radio Archives is using existing metadata fields in archival WAVE files to describe, authenticate, and augment their metadata. Using free or very low-cost tools, alongside well established standards, we will use some of the principles behind W3C's Resource Description Framework (RDF) to choose embedded metadata that is robust, consistent, and surprisingly flexible.