David Cope: A Lifetime Contribution to Artificial Intelligence and Music

David Cope

David Cope is a genius.

Born in 1941 in San Francisco (USA), he developed an extraordinary career as a musician, researcher, composer, author, software engineer and lecturer at USCS (University of California, Santa Cruz).

Furthermore, in my view, David Cope is the greatest contributor to the advancement of artificially composed music.

All companies and researchers that currently work on this field, such as AIVA, Sony CSL, owe a great deal to the incredible theoretical and practical advancements produced by David.

Throughout his career, David created EMI (Experiments in Music Intelligence), published a number of books, journal articles and released a series of records with artificially composed songs.

In this short article, I wish to highlight his work, introduce you to his work and life journey, explain the EMI project, showcase a few songs and provide a summary of his publications.

I honestly hope his work will inspire you as much as it has touch me.

It is a brilliant combination of heart and brain. Art and Science.

Oral History of David Cope

It is always very interesting to listen to David: it is a mixture of humor, history, science, technology, and of course, music.

The passion and curiosity he displays to always dig deeper into music, data and software development is truly touching.

The interview shown below to the Computer History Museum in March 2018 is a wonderful narrative journey through of David’s life. Starting with his childhood, passing through the history  of computers and its connection with music composition, the interview allows the viewer to have a brilliant perspective of the man before behind the sounds and algorithms.

Here it is:

Experiments in Music Intelligence (EMI)

In 1977 David started working at USCS (University of California, Santa Cruz). Later in 1981, he faced a common problem for musicians: composer’s block.

David simply ran out (at least temporarily) of good ideas for compositions. And in order to address this issue, he decided to combine his knowledge of programming with his passion for music to create a program that could quickly and efficiently compose original tracks.

This idea was named EMIExperiments in Music Intelligence.

Using initially an Apple ii e,  David focused on simply finding a more optimal solution to the task of composing songs. After numerous failed attempts, he finally noticed that the secret lied not on developing software not based on “rules“, but rather on “data“.

As he explain in his official site, Experiments in Music Intelligence is grounded in three main principles, which is the base until this day for artificial music composition:

1. Deconstruction: Analyzing music compositions and separating them into parts.
2. Signatures:
Identifying commonality, which signifies and characterizes a style of a genre or composer.
3. Compatibility:
Recombine pieces, patterns and styles to create new original works.

After this, over 11,000 songs were created artificially.

Interestingly, David doesn’t see EMI as a composer. But rather, as a composers’ tool. According to him, “there is nothing magical happening there, it is basically addition and a very simple binary math”.

Music Samples from EMI

As a result from EMI, an incredible amount of songs were created shared with the world.

I have included a few here to showcase what was composed by the software with the intent of resembling specific composers.

Here they are:

1. Beethoven Style

2. Mozart Style

3. Mendelson Style

4. Chopin Style

5. Brahms Style

6. Bach Style

In case you wish to listen to more compositions from EMI, I suggest you visit their YouTube channel or his Spotify channel.

Lecture from David Cope

I always appreciate to see how academics and researchers present their ideas to audiences. The video below shows a lecture of David Cope at UC Santa Cruz.

As the description of the video states, “David Cope will explain why he created his computer program Experiments in Musical Intelligence, how this program works, and how he created over 11,000 music compositions with it. The presentation will include a musical Turing test, a composed-on-the-spot computer composition, and a discussion of how computer composition challenges many on issues of creativity“.

Here it is:

Selected Publications from David Cope

The scientific output of David Cope is highly comprehensive. It is very difficult to include all of his work, so I have comprised here 19 of his publications, ranging from 1976-2009.

The selected publications include conference papers, scientific articles published in journals and books. I have referenced them here on APA style formatting, with the original source link attached to each one of them.

Here they are:

  1. Cope, D. (1976). New music notation. Kendall/Hunt Publishing Company.
  2. Cope, D. (1977). The Mechanics of Listening to Electronic Music. Music Educators Journal, 64(2), 47-51.
  3. Cope, D. (1987). An expert system for computer-assisted composition. Computer Music Journal, 11(4), 30-46.
  4. Cope, D. (1987). Experiments in musical intelligence. In Proceedings of the International Computer Music Conference. San Francisco.
  5. Cope, D. (1989). Experiments in musical intelligence (EMI): Non‐linear linguistic‐based composition. Journal of New Music Research, 18(1-2), 117-139.
  6. Cope, D. (1991). Recombinant music: using the computer to explore musical style. Computer, 24(7), 22-28.
  7. Cope, D. (1991). Computers and musical style (Vol. 6). Oxford: Oxford University Press.
  8. Cope, D. (1992). Computer modeling of musical intelligence in EMI. Computer Music Journal, 16(2), 69-83.
  9. Cope, D., & Mayer, M. J. (1996). Experiments in musical intelligence (Vol. 12). Madison: AR editions.
  10. Cope, D. (1997). Techniques of the contemporary composer.
  11. Cope, D. (1997). The composer’s underscoring environment: CUE. Computer Music Journal, 21(3), 20-37.
  12. Cope, D. (1999). Facing the music: Perspectives on machine-composed music. Leonardo Music Journal, 79-87.
  13. Cope, D. (2000). New directions in music.
  14. Cope, D. (2000). The algorithmic composer (Vol. 16). AR Editions, Inc..
  15. Cope, D. (2003). Computer analysis of musical allusions. Computer Music Journal, 27(1), 11-28.
  16. Cope, D. (2004). Virtual music: computer synthesis of musical style. MIT press.
  17. Cope, D. (2004). A musical learning algorithm. Computer Music Journal, 28(3), 12-27.
  18. Cope, D. (2005). Computer models of musical creativity (p. xi462). Cambridge: MIT Press.
  19. Cope, D. (2009). Hidden structure: music analysis using computers. AR Editions, Inc..

Final Thoughts

David Cope is a man with a life goal. Someone with a vision combined with passion. Someone that searched for years optimal ways to, possibly, reinvent music.

Or at least as we know it.

And this is exactly why his work is so interesting and why it has provided the immense contribution to the development of artificial creativity.

Personally, I do not think he receives the recognition he deserves in the field of AI and creativity. But there is no question that he has made his mark in history.

As I said before, David Cope is truly a genius.