Lupa

Show document Help

A- | A+ | Print
Title:AI and Musical Interpretation
Authors:ID Marx, Wolfgang (Author)
Files:.pdf ZUP_Marx_Wolfgang_2025.pdf (225,01 KB)
MD5: FE12B47C5D8B4B0AAED7568B6F88E2FF
 
Language:English
Work type:Not categorized
Typology:1.16 - Independent Scientific Component Part or a Chapter in a Monograph
Organization:ZUP - University of Primorska Press
Abstract:Generative AI is used to create music more and more. This chapter explores how this phenomenon affects aspects of musical interpretation. The way in which AI generates texts, pictures, music and other media indicates that compared to human creators – it is likely to reduce the range of interpretative variants in its output. This argument is developed in three steps: firstly, a number of examples of AI-generated songs are presented in order to outline what AI can do at the moment. Secondly, the way in which large language models (LLMs) – the underlying systems of many AI generators work is discussed, showing that LLMs have a tendency to standardise their results, while also removing any quirk or idiosyncrasy that human creators would (unconsciously) include. AI outputs are “depersonalised”, as it were. Thirdly, the concept of “model collapse” is introduced. It leads to further homogenisation as AI systems will inevitably include more and more data in their processes that are already AI-created, setting in motion a spiral of self-referentiality that is likely to lead to a collapse of the systems in the long run. While this may not be inevitable it poses enough of a risk to give us food for thought and spend more time assessing the danger posed to interpretative variety by generative AI.
Keywords:AI-generated music, music and model collapse, deepfakes of singers’ voices, AI and musical interpretation
Publication status:Published
Publication version:Version of Record
Publisher:Založbe Univerze na Primorskem
Year of publishing:2025
Number of pages:str. 133-146
PID:20.500.12556/RUP-22270 This link opens in a new window
ISBN:9789612934255
DOI:https://doi.org/10.26493/978-961-293-425-5.133-146 This link opens in a new window
Publication date in RUP:18.12.2025
Views:163
Downloads:2
Metadata:XML DC-XML DC-RDF
:
Copy citation
  
Average score:(0 votes)
Your score:Voting is allowed only for logged in users.
Share:Bookmark and Share


Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.

Record is a part of a monograph

Title:Glasbena interpretacija: med umetniškim in znanstvenim : [znanstvena monografija z mednarodno udeležbo]
Place of publishing:Koper
Publisher:Založba Univerze na Primorskem, Festival Ljubljana
Year of publishing:2025
ISBN:978-961-293-425-5
COBISS.SI-ID:222878467 This link opens in a new window
Collection title:Studia musicologica Labacensia
Collection ISSN:2536-2445

Licences

License:CC BY-NC 4.0, Creative Commons Attribution-NonCommercial 4.0 International
Link:http://creativecommons.org/licenses/by-nc/4.0/
Description:A creative commons license that bans commercial use, but the users don’t have to license their derivative works on the same terms.

Secondary language

Language:Slovenian
Title:Umetna inteligenca in glasbena interpretacija
Abstract:Generativna UI se vse pogosteje uporablja za ustvarjanje glasbe. To poglavje proučuje, kakšen vpliv ima ta pojav na vidike glasbene interpretacije. Glede na način, kako UI ustvarja besedila, slike, glasbo in druge vsebine, je verjetno, da bo v primerjavi s človeškimi ustvarjalci zmanjšala razpon interpretacijskih različic v svojih rezultatih. Ta trditev je razdelana v treh korakih: najprej je predstavljenih več primerov pesmi, ki jih je ustvarila UI, da bi orisali, kaj trenutno zmore. Drugič, opisan je način delovanja velikih jezikovnih modelov (LLM) osnovnih sistemov številnih generatorjev UI, iz katerega je razvidno, da so LLM nagnjeni k standardizaciji svojih rezultatov, pri tem pa odstranijo tudi vse posebnosti ali idiosinkrazije, ki bi jih utegnili (nezavedno) vključiti človeški ustvarjalci. Rezultati UI so tako rekoč »depersonalizirani«. Tretjič, predstavljen je koncept »kolapsa modela«. Ta je posledica še večje homogenizacije, saj bodo sistemi UI v svoje procese neizogibno vključevali vse več podatkov, ki jih je že ustvarila UI, s čimer bo sprožena spirala samoreferenčnosti, ki bo na dolgi rok verjetno pripeljala do propada sistemov. Čeprav to morda ni neizogibno, pa predstavlja dovolj veliko tveganje, da bi se morali zamisliti in več časa posvetiti oceni nevarnosti, ki jo generativna UI pomeni za interpretacijsko raznolikost
Keywords:glasba, generirana z UI, glasba in kolaps modela, globoki ponaredki glasov pevcev, UI in glasbena interpretacija


Comments

Leave comment

You must log in to leave a comment.

Comments (0)
0 - 0 / 0
 
There are no comments!

Back
Logos of partners University of Maribor University of Ljubljana University of Primorska University of Nova Gorica