​INTRODUCTION 

How do artists adjust to the new integration of AI in the Arts? In this world where Artificial Intelligence can produce digital art and take over many other fields of employment, there is a growing fear surrounding the new tool.  Artists fear their jobs will be replaced, and there will be a loss of human authenticity in the Arts. Many are also concerned about AI’s large energy consumption, contributing to the global warming crisis. Lastly, its inherent exploitative nature of using (likely) stolen data to generate images, audios, etc., causes many artists to avoid this medium. My Arts Capstone explores the inquiry, “How can AI benefit a field in the arts without detracting from human authenticity?” This question is addressed by creating five tracks incorporating AI-generated samples. My project weighs the pros and cons of using AI sampling in DJing. Through my use of the tool, without it overtaking my work, I will address issues of human authenticity and ownership of AI art. 

LITERATURE REVIEW 

I looked to DJs and digital music creators as a muse for the style of tracks I wished to produce. For example, I compiled a Spotify playlist of songs from artists such as Crystal Castles, Somewhere Special, and more. These artists do not use AI sampling, however, I wanted to produce human-authentic music similar to their tracks. This would be to demonstrate how it is possible to get a similar product with the use of AI as a tool. I utilized research studies and historical texts to explore the benefits and disadvantages of AI’s use in DJing and digital music creation. For instance, I used the article  “Towards Intelligent Music Production: A Sample-based Approach” to discuss the positives of AI usage. This paper discusses a possible “sample arrangement” titled “Intelligent Music Production”, which enhances human creativity and reduces cost. They created the “MusiComb” prototype, which operates as a sample-generating tool for users that assists human creativity rather than hinders it. It uses metadata such as genre and tempo to generate samples matching the artist’s tracks. This article presents how AI can produce music quickly, is easily accessible for users who are physically disabled or lack transportation, and can be utilized as a tool rather than a replacement for human creations.  

METHODS

My methods of research are literature reviews and archival research. I examined the studies done by coders and practicing DJs, analyzing these new methods of music creation. This was utilized to shape the techniques I used, such as beat matching and transition. It additionally presented how I could best use AI sampling in music, which is through sampling. I used the site Soundtrap to create my tracks and the AI program ElevenLabs to generate the sounds I would sample. 

AUDIENCE & IMPACT 

My project is targeted to appeal to DJ enthusiasts who appreciate the art and are looking to contribute to it with new, advanced methods. I also hope to appeal to artists who fear the rise of AI and its infiltration of artists, especially those in digital fields. I hope my project will accurately communicate the positives and negatives of AI in the Arts to my audience, with my works demonstrating the possibility of both genres' collaboration. Despite my personal opinions on AI, which are primarily negative, I learned that it is possible to incorporate AI into art without encroaching on human authenticity. 

REFERENCES 

Connell, M. (2012). Talking About Old Records: generational musical identity among older people. Popular Music, Suppl.as Time Goes by: Music, Dance and Ageing, 31(2), 261-278. https://doi.org/10.1017/S0261143012000074

George, N. (2004). Hip-hop’s founding fathers speak the truth. That’s the Joint, 45-55.https://bpb-us-e1.wpmucdn.com/sites.psu.edu/dist/f/5180/files/2014/10/FormanNeal-Thats_the_Joint_The_Hip_Hop_Studies_Readerbook.pdf#page=62

Giuliani, Luca, Allegra De Filippo, and Andrea Borghesi. "Towards Intelligent Music Production: A Sample-based Approach." CEUR WORKSHOP PROCEEDINGS. Vol. 3519. CEUR-WS, 2023.

Joo-Wha Hong, Katrin Fischer, Yul Ha, Yilei Zeng, Human, I wrote a song for you: An experiment testing the influence of machines’ attributes on the AI-composed music evaluation, Computers in Human Behavior, Volume 131, 2022,107239, ISSN 0747-5632,https://doi.org/10.1016/j.chb.2022.107239. 

Huang, H. W., Fadli, M., Nugraha, A. K., Lin, C. W., & Cheng, R. G. (2022, November). AI DJ System for Electronic Dance Music. In 2022 International Symposium on Electronics and Smart Devices (ISESD) (pp. 1-6). IEEE. https://ieeexplore.ieee.org/abstract/document/9980591

Katz, M. (2012). Groove music: The art and culture of the hip-hop DJ. Oxford University Press. https://books.google.com/books?hl=en&lr=&id=Zv8JzuFW-l8C&oi=fnd&pg=PP1&dq=founding+dj+artists+articles&ots=2DUs7uae61&sig=9Vl4hhyb9Jfo2fsB61X5o1W8_PY#v=onepage&q&f=true

Len, V. V., & De Bie, T. (2018). From raw audio to a seamless mix: creating an automated DJ system for Drum and Bass. EURASIP Journal on Audio, Speech, and Music Processing, 2018(1), 1-21. https://doi.org/10.1186/s13636-018-0134-8

Nieuwenhuijsen, B. (2023). Sampling in the Age of AI. https://arno.uvt.nl/show.cgi?fid=162563

Olayeni, S. (2023). The impact of artificial intelligence (AI) in the music business industry. https://www.theseus.fi/bitstream/handle/10024/808440/Olayeni_Simeon.pdf?sequence=2&isAllowed=y

Reed, M., Zhao, A., & Kim, H. Audio Reactive AI Generated DJ Visuals. https://www.mattreed.io/projects/djai/djai.pdf

ACKNOWLEDGMENTS 

I would like to thank my faculty advisor, Heather-Erin Bremenstuhl,  and peer mentor, Isabella Pham, for guiding me along my capstone project. I could not have produced this work without their help. I would also like to thank my family and friends for supporting and encouraging me throughout this process.