Too Much Music

How to Define Musical Identity for Humans & AI Systems

September 24, 2023 Alison and Greg Wilder Season 2 Episode 5
How to Define Musical Identity for Humans & AI Systems
Too Much Music
More Info
Too Much Music
How to Define Musical Identity for Humans & AI Systems
Sep 24, 2023 Season 2 Episode 5
Alison and Greg Wilder

 Alison & Greg talk about how we hear, understand, and define the identity of a piece of music. After a bit of a dive into the great music thinkers of the 20th and 21st century and their work on how our minds hear and process music, they jump into the specifics of what musical parameters go into identity, and how we can define and codify those parameters so that they're available to an AI system.

Doctor Body is Alison & Greg's experimental electronic duo.

PatternSonix is Alison & Greg's niche consulting firm focusing on the intersection between sound and AI.

Subscribe on your favorite podcast platform or check us out at

Music, blogging, & more from your hosts: Dr. Greg Wilder | Alison Wilder

Show Notes Transcript

 Alison & Greg talk about how we hear, understand, and define the identity of a piece of music. After a bit of a dive into the great music thinkers of the 20th and 21st century and their work on how our minds hear and process music, they jump into the specifics of what musical parameters go into identity, and how we can define and codify those parameters so that they're available to an AI system.

Doctor Body is Alison & Greg's experimental electronic duo.

PatternSonix is Alison & Greg's niche consulting firm focusing on the intersection between sound and AI.

Subscribe on your favorite podcast platform or check us out at

Music, blogging, & more from your hosts: Dr. Greg Wilder | Alison Wilder

Alison Wilder (00:04.171)
Welcome to Too Much Music. And hey, what's going on? I'm Alison. This is Greg, yeah. And we're here to talk to you today about musical identity. But first, how about our own identities? Let's talk about those. How you doing, Greg? What's going on in your world?

Dr. Greg Wilder (00:07.61)

Dr. Greg Wilder (00:11.827)
Uh-huh. I'm Greg.

Dr. Greg Wilder (00:26.042)
I'm doing great. I've just undertaken a tremendous studio renovation. So lots of new instruments, new wiring, new, you know, so my studio is in process and that's a very fun thing. It is chilly now. So it's the end of September in the mountains of New England. And so it's like I woke up in the, it was in the thirties.

Alison Wilder (00:54.751)

Dr. Greg Wilder (00:55.85)
Yeah, I love it. It smells great, feels great. It's like perfect outdoor working weather for me. So, yeah, I'm happy. Very happy.

Alison Wilder (01:02.171)
Yeah, and I know you have large piles of wood that you're processing personally, which I find...

Dr. Greg Wilder (01:08.414)
It's the trees that grew on the land will warm my cabin in the winter. How poetic is that? Yeah. Hmm. But it's true and it's kind of beautiful.

Alison Wilder (01:12.971)
How quaint. Yeah. It is, it's also beautiful to have kiln dried firewood dumped outside your shed. Ha ha.

Dr. Greg Wilder (01:26.518)
And that is where you are. Having it hand delivered, perfectly seasoned, stacked and yeah.

Alison Wilder (01:28.923)

Alison Wilder (01:32.759)
Oh man. Yeah, you just look at it and it just lights on fire. It just ignites. Yeah, it is nice to get into the cooler weather a little bit and yeah. I'm personally feeling really peppy and happy today because I stopped my insane coffee experiment. I decided that I was gonna try to be without caffeine and see what that was like.

Dr. Greg Wilder (02:00.578)
How'd that work out for you? No, I'm just, I'm like, mm, that sucks, let me.

Alison Wilder (02:02.535)
Yeah, it sucks. I would say the experiment, you know, rendered a very firm result and it's that, you know, I probably should have less caffeine, but not no caffeine.

Dr. Greg Wilder (02:16.07)
Yeah. You inspired me to have less. And so, yeah, and I think it is better. But there's a rough period transition.

Alison Wilder (02:22.489)
Yeah, but-

Boy, it was bad. I did cold turkey for two weeks, and I swear, our last two episodes, my eyes were just glazed over. I could barely talk, I could barely think. Like trying to read articles, and I just kept reading the same sentence over and over again. Turns out coffee's important, so let's... Thanks, coffee.

Dr. Greg Wilder (02:36.213)

Dr. Greg Wilder (02:41.85)

I had someone who watched the podcast come back and say, is Alison high? And I said, no, absolutely not. It's nine o'clock in the morning. What are you talking about? She just stopped coffee. That's all.

Alison Wilder (02:56.765)
Seriously, somebody noticed that?

Dr. Greg Wilder (02:58.666)
Yeah. Oh yeah. You look, you looked a little, a little out of it. It's okay. We love you. It's fine.

Alison Wilder (03:03.787)
I'm sorry everyone, I'm okay.

Oh, it was just an experiment. Yeah, I'm just on.

Dr. Greg Wilder (03:11.894)
But you have an espresso machine too. You have like a one button press espresso machine. So you should use it. Well.

Alison Wilder (03:17.651)
Dude, looking at that every day for two weeks and not turning it on was like torture. It was truly awful. Yeah, no. Yeah, just a lower dose, that's all I needed. And everything's fine now, I'm back. So yeah, and I'm glad because we have a really fun topic today that's near and dear to both of our hearts and we get to do a little bit of geekery, which I know both of us love. And.

Dr. Greg Wilder (03:23.542)
Oh my god.

Dr. Greg Wilder (03:44.078)
power for the course.

Alison Wilder (03:46.18)
Yeah, so we're gonna talk about musical identity. What makes it, what is it? How do you see it? How do you know it? Do you know it when you see it? Ha ha ha.

Dr. Greg Wilder (03:53.262)
We all know it when we see it. We all know it. So maybe I can start and just, um, you know, so I'm a composer. Um, and I spent a lot of years, more than a decade, you know, having mentorships with great composers. Uh, that's what you do. And it's almost always a one-on-one situation. Um, and the thing that happened in most of those, uh,

meetings is that there was a lot and this happened in performance. I'm a pianist as well. And this happened in my piano lessons, too, which is to say people would talk about music in these very metaphorical, lofty, very intangible terms, right? They would leave the analysis for music theory class, so to speak, and want to talk about the music that I was working on.

and the specifics of it in abstract terms, cause and effect, possibility, just sort of floaty cloud type language, which is sort of an old school, my teachers were from a different generation and it was sort of an old school, maybe romantic approach to performance and musical thinking. And I think it has.

Alison Wilder (05:14.991)
Interesting, a romantic approach even though they were hardcore 20th century composers.

Dr. Greg Wilder (05:19.346)
Yeah, well, but as we know, the 20th century sort of fell into two camps, right? There were some people who wanted to break with the past and wanted to look very hard at just the math sort of thing. But not just the math, but a lot put that at the forefront. And then there were others who wanted to adopt that language or, you know, be open to those new possibilities that the 20th century brought, but remain in a romantic spirit. And I kind of sided with those folks.

Alison Wilder (05:47.911)
And who are the, just so our listeners will have a more specific idea of what you're talking about, who are the composers you're thinking of, maybe a couple of composers in each camp that people might know? So, Stackehausen.

Dr. Greg Wilder (06:01.118)
Oh, so Pierre Boulez, you know, for example, he's sort of the, yep. Those are folks who are on the break with the past. We need to start again. Uh, the world has changed and this happened a lot largely because of World War Two, we can, we're not going to get in the history of this, but after World War Two, people did have a major cultural and psychological reset that they needed to make and they felt like maybe the, um, the excessiveness of pre-war romanticism and expressionism.

in Europe, especially may have been part of what led to. The atrocities, right? So, um, so they wanted to pull that back and get control over it. And there was also the fact that this was happening. This was starting to move into the university music was. And, uh, so they needed to justify and sort of scientificify. I don't know where it is. Science of fi. Yeah. Science of the study. Scientific. I like that. That's okay.

Alison Wilder (06:51.409)
Science-ify? Scientificate? I think it's scientificate. I think it's the word.

Dr. Greg Wilder (06:58.018)
They wanted to scientificate it and make it a rational study. Um, and then there were others who just, you know, refused to do that or felt that was not the right way to go. And we're, and well, I studied with George Rockberg that he was, I mean, I studied with many people, but he was maybe the person I studied with the longest. Um, I was with him until his death.

Alison Wilder (07:08.335)
And who are some of those people?

Uh huh.

Alison Wilder (07:18.387)
Considered the father of postmodernism and music. Yeah. Yeah, right. He came first. So Berio would be another example of someone who's more on the romantic side, more writing from the heart, I guess, if you wanna trivialize it a little bit.

Dr. Greg Wilder (07:22.73)
Yeah, before Barrio. I always like to point out he's European, so he got the

Dr. Greg Wilder (07:36.542)
Well, or looking to the past to say, you know, you can't break from the past. You can't just decide that Mozart didn't exist and that doesn't have an influence on us today. And that didn't mean anything to our culture or to our psyche because it does. Right. And so.

Alison Wilder (07:48.643)
Okay. And so the interesting thing that Rockberg did, and this is directly tied, I think, in a really interesting way to musical identity, is that he was maybe the first person, at least the first known composer, to really just grab bits of old music and put them in his music. So he would put Bach a lot, Mozart, the great composers of the classical and Baroque and Romantic era, just...

pop them directly into his scores, right?

Dr. Greg Wilder (08:18.922)
And he would recontextualize them. So you'd have a piece that I have come to play a lot is Nachbach. This is one of his works for piano, which after Bach, right? And the idea there is that there are fragments of Bach flying around inside this very, very modern other worldly texture. And so you, it's a, it's a little psychedelic, I mean, in a way, right. Um, kind of.

Alison Wilder (08:26.591)
That means after Bach. Yeah.

Alison Wilder (08:43.464)

Dr. Greg Wilder (08:45.29)
But the idea, and then Jim George would also later, he would compose in the style of these composers and around their music even. So that was never clear what was him, what was them, maybe, or at least it's hard to tell. And you can fool trained musicians with this stuff. But ultimately that distilled, I think, into an acceptance of, and what we see today, the influence that had is it was an acceptance of bringing, keeping the past alive in a modern setting.

I mean, that's sort of the, that's postmodernism and that's what his legacy, I think, in part is.

Alison Wilder (09:20.511)
So Rockberg, I know, is, you know, I would say your primary composition mentor in your life. He's the person who seems to have had the most drastic effect on you as a composer and as a person who thinks about music. And he obviously had a very deep relationship with musical influence and identity. So you, I think that was probably one of the things that attracted you to him.

as a teacher and as a mentor. So how did that go from, oh, I'm studying with this great person and I'm what, you know, you were young when you started studying with him. How did that go from that to sort of how you incorporated that into your work, both musical and technical work?

Dr. Greg Wilder (10:12.63)
Well, boy, that's a huge, that's a, I mean, that's, yeah. So, well, yeah, I have a, I literally have outlines to write a book about some of this actually. So it's a big topic. I don't know how to summarize it probably very quickly. Suffice it to say, the idea, so the fact that music to me, and this is very relevant to identity and to what we're gonna get into about.

Alison Wilder (10:17.472)
Tell me about your life's work, Greg.

Dr. Greg Wilder (10:42.018)
codifying this, that music to me is an act. It's something that a person does and can offer to others. That meant that I, I mean, this is why I studied piano. I needed, I wanted to be a composer, but I needed to study piano to have a mastery of an instrument to understand what it took to make sound out of a essentially percussion instrument to make it music that came out of it.

to understand how other people throughout the years had treated that instrument to make music happen with it so that I could then turn around and write music that would also do that, that would contain that not magic sauce, but now I'm going to start getting metaphysical on us here. And when I say metaphysical, I don't, I mean metaphorical, but George saw it as metaphysical too, which is to say he saw it as like a life and death struggle.

Alison Wilder (11:29.579)

Dr. Greg Wilder (11:38.442)
I mean, he has a book of essays called The Aesthetics of Survival, and he's talking about how you can't, you know, there's many things he's talking about, but he's saying that survival of culture to him was essential to the survival of humanity, basically, and that we can't lose the past. And that music has to be a thing that we make, that we share, that has deep, deep meaning, culturally, historically, and to us today.

Alison Wilder (11:38.815)

Dr. Greg Wilder (12:07.434)
And that it can't just be, it shouldn't just be an experience that just passes over us and we just sort of, you know, let it flow and then we turn it off. Okay. It has to be something that we engage in. So, and now I would like to turn this around because you're, you're background, you have a background in composition too, and, and performance, but I want to bet you did a lot of theory work. So I want to ask you, did this stuff come up in theory? I mean, did, you know, I've never met a theorist, maybe one who like,

would talk about music in these terms of the cosmic zeitgeist and the flow of ideas that are bigger than a person, that kind of thing. How did that work in theory? Did you find that happening?

Alison Wilder (12:47.032)
Yeah. No, no, theorists don't do that. It's not that they don't think that way. It's that they're not interested typically in writing or in that way. So you might have conversations with theorists, even great ones, you know, that kind of sound the same as the conversations you might have had with George. But when you read their work, well,

George's work was writing the music down. When he wrote the music down, all of those concepts didn't matter, right, in a way, because this is his work that he made. And for theorists, the equivalent of that is what they write about how music works, right? So I would say it's similar in that way. So they might think that way, but it doesn't matter, because they're certainly not gonna put that in their book, because theorists are more interested in

Dr. Greg Wilder (13:23.25)
Mm. Yep.

Dr. Greg Wilder (13:38.626)

Alison Wilder (13:42.911)
getting into the technical details of the workings, the mechanics of music. And I think one of the most interesting things that happened on that front, and the thing that really has inspired me in my work, in my thinking, for a very long time, is the turn from just looking at the music and looking at the score and trying to parse that, which is really interesting, and there's a lot you can do there, and I've done a lot of that,

Going from that to turning to how the mind processes music and how that affects the identity of that music and what it is and what it means and how it works, I think is maybe that's one of my top topics. It's a hot topic for me. I think that's like the most interesting thing there is to look at. Yeah, so maybe we could just talk about that and how that has worked.

Dr. Greg Wilder (14:18.89)

Dr. Greg Wilder (14:25.814)

Dr. Greg Wilder (14:32.142)

Alison Wilder (14:42.899)
We talked about the change from pre-World War II to post-World War II composition, which was just a massive overhaul for everything. Well, basically, everything cultural on Earth was overhauled by that, but certainly music. And music theory as a discipline took a little bit longer to come into its own. And I would say...

Dr. Greg Wilder (14:58.174)

Alison Wilder (15:11.063)
One of the major developments that made that happen was, you know, was this turn to how the mind processes music. Oh, my dog is interested in moving around right now. Something is happening and Peanut needs to know. Yeah, so, go ahead.

Dr. Greg Wilder (15:22.026)
That's okay. She can get in there. That's all right. So, well.

I was going to say we can go back to the fifties and sixties when this sort of theory sort of began, but maybe more, we bring it a little closer to home around the, around the eighties, right?

Alison Wilder (15:42.235)
Yeah, well, in the early 80s, I was born. It's important. The modern Republican Party was born. MTV was born. And a generative theory of tonal music was born. This was a super, yeah, Greg knows this text very well. This was a super important turning point for music theory and I would say for just...

Dr. Greg Wilder (15:48.31)
Ha ha

Dr. Greg Wilder (15:52.078)
Mm-hmm. Ha ha.

Dr. Greg Wilder (15:59.65)
Oh my god.

Alison Wilder (16:10.719)
thinking about music in general and studying it in a more rigorous and scientific way, which as Greg said, you know, people started trying to think about it that way earlier, but I would say the eighties was when it actually started to happen. The GTTM.

Dr. Greg Wilder (16:25.134)
People were looking to codify meaning. That's how it started. They, they, people were, you know, they understood the structure of music with theory, plenty of tools for that. And then some people started to say, but how do we capture meaning? How do we codify meaning? That eventually led to some of this work.

Alison Wilder (16:41.403)
Yeah, yeah, and the great book, Emotion and Meaning of Music by Leonard Meyer kind of kicked that off earlier. And by the 80s, people were trying to come up with testable, verifiable theories about that. And I think one of the first really successful ones and probably one of the most influential moments for music theory as a discipline was GTTM, or Generative Theory of Tonal Music. So this was written by

Lairdahl and Jackendorf, Fred Lairdahl and Ray Jackendorf, and I think 83, probably. Somebody can correct me.

Dr. Greg Wilder (17:17.614)
I thought 82. One of the interesting things about them is that Fred Lairdahl is a composer, not a theorist, and Ray Jakendoff is a linguist, not a musician. That's really, yeah.

Alison Wilder (17:29.007)
Yeah, so they got together, and they were both really into Gestalt psychology concepts as well. They read a lot of that. And, you know, I'm not a linguist, so hopefully I'm not getting this super wrong, but they were heavily, Jackendorf was heavily influenced by Chomsky, as my understanding, in generative linguistics. Yes. And yeah, so they were looking at...

Dr. Greg Wilder (17:49.014)
They were at Columbia, right?

Alison Wilder (17:57.691)
music as a mentally constructed entity. And this is really the first time that music theory, I would say becomes a branch, honestly, of cognitive, what they called cognitive science at the time, or maybe theoretical psychology. So they weren't really, you know, you'll see them analyze scores, but what they're really trying to do is analyze how our minds process music. So they were looking at components that

Dr. Greg Wilder (18:12.247)

Alison Wilder (18:26.195)
that influence components of the music that influence our cognition of it. So they looked at things like the grouping structure that they called. So when you look at their analysis, these are kind of the bits of the analysis that you see, or if you try to do a GTTM style analysis, this is what you do. So you create structures that are groups. And for them, this is a generic term, right? So, and music can have groups on many levels, but it's just any salient musical segments. So it could be,

a motive, it could be a phrase, it could be a section, a theme group. So we need to figure out how to segment this stuff. Then they looked at metrical structure, which is just simply patterns of strong and weak beats. So how do we tap our toes to this, things like that. What's the pulse that we feel. And then they would do what they call time span reductions. And these are kind of the links between the grouping structures and the pitch and the rhythm.

And then something very interesting that I find really fascinating that they would do is they would look at what they called prolongation. And they would try to reduce the prolongation. And this is just patterns of what we perceive as tension and relaxation. So this maybe coincides with grouping structures and maybe different, just depending on the way the music is written. There are a lot of criticisms of this work. I mean, I would guess.

hundreds of papers have criticized the GTTM rightfully.

Dr. Greg Wilder (19:54.85)
Mm-hmm. And yet, and yet when I came up and took theory, there was a lot of discussion about tension and release and these very similar, these much, these principles of grouping and structures and such.

Alison Wilder (20:08.103)
Well, that's the thing. There's a lot to criticize, but they turned out to be in a lot of ways, really on the right, I would say on the right track. You know, they were only, their theory, they wanted it to be very scientific. So as with a lot of, you know, especially like, if you look at behavioral studies about music, for example, they're very limited because in order to...

create a viable scientific experiment, you have to isolate variables, blah, blah. All these things are very hard in music. So like that, you know, Lerda and Jack and Dolph did this on very simple monophonic melodies, basically. They were really only looking at melodies, which like obviously is a tiny piece of what we hear when we hear music. They didn't have anything to do with timbre. They, you know, they...

Dr. Greg Wilder (20:46.635)

Dr. Greg Wilder (20:54.027)

Alison Wilder (21:01.039)
made a lot of claims about universality, which have actually not turned out to be totally wrong, which is interesting that there have been lots of actual experiments to test these things in the 40 years since this was written. And, but they, you know, it's like non-falsifiable claims, very non-scientific stuff. And also they only worked on the Western music, tonal music canon.

Dr. Greg Wilder (21:07.223)

Dr. Greg Wilder (21:24.95)
Now it's as an anecdote here, so I, we can get into this, but I came to the GTTM as I began my own process, trying to deep dive into building Cleo, which we talked about last episode. Uh, and I needed to understand how I might bridge musical mood and meaning into analysis and such like this, cause I wanted to codify it. So they, I learned about the GTTM after graduate school this way. Um, but as I understand it.

from people I know who know Fred Leradal. He was really trying to find ways to justify the romantic side of composition in a way. Like in other words, he was trying to say, we can't process, I think, this was his personal motivation, probably where a lot of the intuition came from, but he was saying, we can't process a lot of this extremely mathematically derived, break from the past kind of music. Our brains can't do it, our ears can't do it.

And this is why, kind of, right? That's where a lot of that motivation was, secretly.

Alison Wilder (22:28.811)

And I think he was very, I think he was just flat wrong about that. I think that was a personal preference that, you know, that seeped into his work. And it turned out to point him at some really interesting things that actually are pretty universal. But, you know, to say we can't understand, I do think there are probably, I mean, we know there are patterns that are too complex for us to process, but I don't actually think it's because

Dr. Greg Wilder (22:39.846)

Dr. Greg Wilder (22:55.863)

Alison Wilder (22:59.751)
We can't, I think it's because we don't for the most part. I'm sure there's a hard limit. I don't know anything about what that would be, but this kind of brings me to, you know, another really interesting person in this same lineage and history. And we'll skip around a little bit because it's directly related to this, which is David Huron. And we can talk a little bit more about his book and his theories. He wrote a book called Sweet Anticipation. This was much later. So.
in 2008,  so significantly later than the GTTM, certainly influenced by it. But.

Dr. Greg Wilder (25:56.138)
And the title, sweet anticipation, is referring to the idea of that tension and release that we were just referring to, right? And how that works.

Alison Wilder (26:04.079)
Yeah, he had a really novel theory of, and has actually continued to develop this and gone even beyond music to develop an entire very compelling theory that is called the ITPRA theory. It's a theory of expectation that, you know, that kind of has been influential and deals with how we interact with the world in general. So ITPRA meaning imagination, tension, back to tension.

prediction, reaction, and appraisal. So he has a theory of how those, that series of things of mental structures is how we process everything that's incoming. And it's especially.

Dr. Greg Wilder (26:46.562)
So we're playing a game as we experience something like music. We're playing a game with ourselves where we hear something, we process, we, we feel tension, we predict what we think is going to happen. Then we appraise what really did happen and feel good or bad about the result.

Alison Wilder (27:03.483)
Yeah, yeah. And so when we're enjoying something, you know, whether it be a film or a piece of music, he would say that we're not just enjoying it, you know, just because it's great or beautiful or intense or whatever. What we're actually enjoying about it is our own predictions that we've made in our subconscious mind being wrong or right.

And the way in which that works, you know, we always like them to be right. That's always a nice feeling, like, oh, I was right about that. You know, our brains like that. But it turns out we also like to be wrong, but only in very particular ways, right? There are ways that you can craft a piece of music that bucks our predictions and is great, and we like that. Or you can craft a piece of music that bucks our predictions and we hate it. But back...

Back to Lairdahl and why I think that David Huron has a much more interesting take on the tension release expectation issue and more universal is that David Huron talked about one of his major, I think a major breakthrough in thinking about music was what he called statistical learning. So what he said is our minds are forming,

what's essentially a statistical analysis as we listen, you know, our entire lives. We're forming a basis for how we think music works and it comes from what we hear, right? And yes, there are some universals, but really we're creating this, you know, in our own minds throughout our entire lives.

Dr. Greg Wilder (28:46.518)
So we're being trained, we're being trained like an LLM, sort of, from the beginning. We hear Sesame Street, whatever, we hear early music as a child, and we're getting trained on how music functions and how it works and what we expect it to do and how, right? And in some detailed ways, even though we don't understand, necessarily, the theoretical structure of this, we're getting, much like spoken language, we accumulate that intuition and knowledge through training.

Alison Wilder (28:51.315)

Alison Wilder (29:15.367)
Yep, so we intuit what's going to happen in music that is at least somewhat in the realm of things that we're familiar with. And so I think that one of the problems with the mathier 20th century music, let's say starting with Schoenberg, and I mean, Boulez is a great example, Boulez is very, very difficult music to parse.

Dr. Greg Wilder (29:16.148)

Dr. Greg Wilder (29:25.623)

Alison Wilder (29:45.031)
as you listen. But I would say it's not, you know, unlike what L'Erdal seems to have thought, it's not that, you know, we're not made to do that or that we can't do that. It's that we haven't done it. You know, most of us didn't spend our childhood listening to Schoenberg and Boulez. We listen to Sesame Street. We listen to stuff that, you know, it's complex because it's stuff that was influenced both by, you know, Western classical music and also...

you know, music in the blues tradition, right? So, and there are some related things there, but you know, what we certainly didn't grow up listening to is, you know, structure. Like we can't, yeah, what if we had? There probably are. I mean, I bet a lot of, you know, I bet a lot of great conductors started listening to that early, those conductors are great listeners.

Dr. Greg Wilder (30:26.41)
And so what if we had and are there people who have? Are there people who have? Can we?

Dr. Greg Wilder (30:38.802)
I know that my experience was as a graduate student in my teens and twenties, like I became aware of this music and spent a lot of time with this music because I knew it was important. I also knew it didn't come easily. And I played it, I listened to it, studied it, and in doing so, I definitely began to find the... I found a new way of listening, is I guess how I would say it. I found an ebb and flow with it.

felt very feels very natural today even. And in fact, in a way it's sort of warped my hearing so that when I compose something that's a little bit out there even just a little bit, a lot of people find it just I'm not sure how to follow it. I don't know what's going on to me. I mean, it's almost silly how simple the structure might be or how clear I've made it. But it's not always obvious. But I think my mind has warped a little bit to be a little more open from all that.

Alison Wilder (31:11.431)
Yeah. Oh, me too.

Alison Wilder (31:22.178)

Dr. Greg Wilder (31:37.47)
Years and years of deep, intense listening and playing.

Alison Wilder (31:40.647)
Yeah, well, I've had the same experience. You know, I also spend a lot of time with pretty complex 20th century music, both listening and with a score. For me, the score, working with the scores helped a lot with my listening, you know, because I was able to parse them that way and then, you know, use that understanding to listen. So anyway, the educated listener is a very interesting topic actually.

Dr. Greg Wilder (31:59.164)

Dr. Greg Wilder (32:07.078)

Alison Wilder (32:07.827)
because it does 100% there, you know, probably dozens if not hundreds of studies at this point, behavioral studies mostly proving that educated musicians listen differently. You know, they're just, you hear things differently. You hear finer segments, right? You hear more segments if you know what's going on.

Dr. Greg Wilder (32:30.646)
Well, and how could that not be the case? And yet culturally at this moment in time, it is elitism and it's very frowned upon. Right. This is not something we should ever say or talk about yet. That's why we should do a podcast episode on it, but later maybe, but, um, well, music elitism can, behaving in an elitist way is never good for anyone. But the idea that some music does take a lot of work and is a lot of, it takes a lot of effort.

Alison Wilder (32:45.885)
No, we should.

Dr. Greg Wilder (33:00.774)
is not, that's just true. And there's nothing wrong with that, but it does mean that listenership will be limited and certainly, especially given the way we take in music today, but.

Alison Wilder (33:14.443)
Right, stick the white things in your ears and go about your business. Right? This is exactly the opposite of what Rockberg wanted to see happen.

Dr. Greg Wilder (33:17.934)
Sure, the soundtrack of your life.

Dr. Greg Wilder (33:26.646)
sure, or he felt was essential to the survival of culture. And, you know, so he was a Jewish man, you know, who fought in World War II. This is worth mentioning that, you know, like on the American side as an American soldier. And this is important because he lived that struggle very directly, that particular, the death of, the attempting attempt to extinguish a culture.

Alison Wilder (33:57.083)
Yeah. His culture. Yeah.

Dr. Greg Wilder (33:58.282)
And a right, yeah, his culture. Um, so he certainly had a very different kind of, uh, relationship to that survival that would, than we would today. Um, any.

Alison Wilder (34:08.767)

Alison Wilder (34:12.379)
Yeah, that really changes the idea of the aesthetics of survival when you know that about him. He's just such a... I never met him, but I wish I had. And in my life, he's just such an interesting character. I've gotten to hear so many great stories about him from you over the years. And studied him in school. He's just an absolutely incredible composer. But yeah, that experience must have been... You don't come back from that. That changes you.

in fundamental ways that bleed into everything you do.

Dr. Greg Wilder (34:45.73)
There's a colleague of mine from Eastman who just wrote an autobiography, sorry, not an autobiography, a biography of him. But that's not quite right. It's specifically about his change, cultural and artistic aesthetic change from before the war to after the war because it was significant. And in fact, she interviewed me. I'm in the book actually having a postmortem conversation with him, which is pretty cool at the end. But.

In the book, though, she talks about how his handwriting, she was going through all of his documents, and his handwriting before the war was one way, and then after, it was all different. And the way he wrote music was different. Everything about him was different. I mean, that can't be an isolated case, and that's not something to be lost or forgotten. Just because we're not experiencing that today, that's something that happened to humanity that I think is worth.

but we're kind of getting off topic here. But, but.

Alison Wilder (35:47.615)
Whew, that's a big, that's a big, we need to maybe just sit here and breathe for a minute after talking, that's a very intense topic. But yeah, let's go back to, you know, circling back to these kind of pivotal people and thinkers and lines of thought.

Dr. Greg Wilder (35:50.354)
Yeah, it's well, so.

Dr. Greg Wilder (36:07.274)
And people who are trying to codify meaning in music or emotional reaction or cognition, really, how can we, what's our limit? What, how do we parse it? Yeah. Not just how does music, not just how does music get structured, but how do we pull it back apart when we hear it and experience it?

Alison Wilder (36:17.776)
Yeah, so.

Alison Wilder (36:24.739)
which let's be sure to also talk about how that, how we think, you know, how does that relate to musical identity, which is our actual topic here. So let's circle back around. Why are we always circling back? I'm tired, I'm gonna stop saying that. Sorry, everybody. This is what happens. Yeah, I like that.

Dr. Greg Wilder (36:35.95)

Dr. Greg Wilder (36:46.506)
Reverse the train. No, go forward picking up another path. I don't know. I lost it.

Alison Wilder (36:55.823)
Our brains have been turned into ooze by business jargon. Let's circle back to a little dude called Eugene Narmor. So this is kind of the third, what I would consider, big stake in the ground of music analysis or thinking specifically about music. And this was, he came along before and was probably very influential on David Huron.

Dr. Greg Wilder (37:01.812)
Mm. Mm-hmm.

Alison Wilder (37:23.583)
And he came up with something called the implication realization model, which also like the GTTM, how many Ts are in that? GTTM has Gestalt underpinnings. So I know that's something that you have been super influenced by. Maybe you can talk a little bit about the implication realization model.

Dr. Greg Wilder (37:47.45)
Yes. So I found the GTTM and then quickly found Narmore through musician friends, not theorists, but through musician friends because they were so, well, they were so attracted to what he, what these people were writing about and the way they were approaching music theory because it's so intuitively clear to a practicing musician. And anyway, so it was very interesting. And so I went to Penn and met Narmore and talked to him about my work and wanted to codify it. Now,

Alison Wilder (37:55.403)
Hm. That's weird.

Dr. Greg Wilder (38:17.442)
This was in the early 2000s when I was doing my work. So only about 10 years after he had published this stuff, which it has since been clinically tested and there's a lot of work that's been done on his work. His work has had just, as you're saying, a wave of influence on everything that's come before or after. But the idea simply is, it's the sort of a very mathy version of the Huron concept, which is you can have an event, a musical event, you can have another musical event.

Alison Wilder (38:22.591)

Dr. Greg Wilder (38:46.178)
Those two, this is the watered down version, those two events in proximity now create an expectation, an implication of a third event, which we haven't experienced yet. So when we hear that third event, it's realized, and the question becomes, to mix Huron and Narmore, how much did we, how satisfied were we with that resolution or with that realization? How dissatisfied?

Alison Wilder (39:14.047)

Dr. Greg Wilder (39:15.106)
How much did we expect it? Did we predict it? How accurate were we? If we are, and here's the short, here's the super short of the longterm of this. If I keep providing these events that thwart your predictions, if I just constantly thwart your predictions and create randomized stochastic events, you will be bored. You will not hear music. You will not remember what happened 15 seconds ago. You will just have an experience.

Alison Wilder (39:41.327)

Dr. Greg Wilder (39:45.034)
and you can feel how you feel about it. If instead.

Alison Wilder (39:47.423)
So there'll be no musical identity to that whatsoever. Okay.

Dr. Greg Wilder (39:50.858)
That's right. The identity kind of goes to zero because you can't get there. You can't figure out what it is and you stop trying. If instead though, if I go to the other extreme and I just keep giving you exactly what you predict all the time, every step of the way, you will be bored again. Because there's nothing interesting here. You're just hearing a scale or whatever. You're just hearing the most obvious continuation of the previous events. And you develop, um, what I.

what I devised, you develop an equity to that process and you need it to surprise you at some point. That's important. You need it to surprise you in that sweet anticipation kind of way. You need to be able to predict how to be wrong in just enough of a way that you're engaged and you want to know who done it. You want to know what's next. You want to know the answer, right? So

Alison Wilder (40:41.407)
So hold on, you said something really interesting in there. Well, it was all super interesting, obviously. But you said, but you said, oops, you said develop an equity. So I think what you mean by that is like you put equity in a home, right? You're putting, yeah.

Dr. Greg Wilder (40:46.594)
Thank you. Of course it was.

Dr. Greg Wilder (41:01.474)
Literally, you're just putting money in the bank account. You're like, you're like, you make a prediction. You're right. You're like, awesome. You get a little coin in the I'm right jar. You do it again. I'm right jar. I do it again. I'm right. Oh, no, I'm not right. And that was really surprised me. And the more equity you build up, the more impact that surprise will have typically. Right. So, yeah, this is this is the underpinnings of a lot of what I codified and built into Clio in terms of its

analysis of musical language.

Alison Wilder (41:34.184)
So maybe you could give some, let's talk with some specificity a little bit about that, about the different parameters. So obviously one of the things that's difficult about studying music in any sort of scientific way is that there are lots and lots of parameters when you're listening to music. So we can talk about melody, which is essentially frequency and order of frequency and.

how long and short each frequency is. Yeah, time series data, if you wanna get real nerdy about it. You can talk about, just about rhythm, just about the onset, event onsets, how long they last and how they relate to each other, how, you know, metrical grouping, like L'Erdal talked about. You can talk about timbre, which is something that's super hard.

Dr. Greg Wilder (42:04.258)
time series data.

Dr. Greg Wilder (42:18.39)

Alison Wilder (42:28.627)
And something that the lab I used to be in at McGill has really focused on over the years, which is how do you codify timbre, right? Like how do you, how does that play into all of these? Cause these are all kind of simplistic, you know, we need these very basic, here we have a melody, you know, it has eight notes, they're in this order, it's super obvious to analyze something like that. But as soon as you add timbre or orchestration, like that goes out the window, right?

Dr. Greg Wilder (42:37.314)

Dr. Greg Wilder (42:56.802)
Well, harmony, I would add harmony too. So yeah, polyphony and harmony is melody, but it's coincidental melody. And so that will change whatever you perceive as that top melody, it can change very directly its meaning. So.

Alison Wilder (42:59.651)
Yeah, yeah, not only that. Yeah.

Alison Wilder (43:15.379)
So it's just layer on layer on layer on layer, right? And this is what has made it very hard.

Dr. Greg Wilder (43:18.078)
Yep. And in fact, you know, like every one of these could be a book because like you just mentioned, rhythm sort of, okay, we have pulses and all that and we want to, but there are pulse patterns that can be ambiguous, that could be many different, that could be set up and structured in many different ways that would all be justifiable. And then there's a gothic accent, which is a rhythmical accent in a melody that will create the illusion or the perception of an accent.

that isn't accented, an accented event that has more importance, but it isn't in any other way except that rhythmically it's been accented, right? So leaned on in some way. Yeah. So these interact in ways, it gets very complicated very quickly because they interact and they go very deep into how they actually present themselves in any music.

Alison Wilder (43:56.143)
leaned on, yeah. Yeah.

Alison Wilder (44:12.123)
So let's take this idea, and this is something we worked on a lot in Clio, the idea of the perceptual pulse pattern. So a lot of times, people in general music land, you go on YouTube and you start listening and people talk about music, they'll talk to you about BPMs, right? Which is Beats for Minutes.

Dr. Greg Wilder (44:33.519)
Yes. Uh huh.

Alison Wilder (44:36.343)
And it's an easy metric to talk about. And typically people think that that's how tempo works. It's BPMs, right? And that's, I mean, we're laughing because it's just, it's very oversimplified because there is, yeah, it's a number that has something to do with how we perceive the pulse, usually, probably, if it's calculated well.

Dr. Greg Wilder (44:52.458)
Yeah, it's a number. No.

Alison Wilder (45:06.047)
BPM, sorry, it means beats per minute. So that means the idea is if you're tapping your toe to, you know, or if you're clapping along with a piece of music, it's going at a certain number of beats or pulses per minute. And that's just a convenient number that we can use to estimate tempo. But talk about perceptual pulse patterns a little bit because that's a much more interesting way of dealing with rhythm.

Dr. Greg Wilder (45:31.274)
So yeah.

Alison Wilder (45:34.095)
in the context of these implication and realization type models, right?

Dr. Greg Wilder (45:38.71)
want to do that and then I want to jump back to George and the metaphysics of this actually.

Alison Wilder (45:43.615)
Ooh, oh my God, you can go from perceptual pulse patterns to metaphysics? All right.

Dr. Greg Wilder (45:47.102)
Oh, yeah. Absolutely. So I created that phrase, perpetual pulse patterns, it's in a provisional pattern that we have. I did. And the reason I did that, I created that was because as the more we did this work, I realized that the idea of rhythm, whatever that means, you know, there are so what is the bass drum doing? What is that guitar pattern doing? What is the vocalist doing? Rhythm. We can write it down. That's fine.

Alison Wilder (45:53.031)
You can't even say it, you didn't create it. Ha ha ha!

Dr. Greg Wilder (46:13.218)
but much more of an impact in terms of how we feel about the mood and the experience of music that we're hearing is that perpetual, I'm sorry, perceptual pulse pattern. How are we feeling a pulse? So we can have very slow music that feels very fast on the surface. So we can have a very slow moving underlying pulse that has a lot of activity over top that feels very fast. And that still may feel slow.

Alison Wilder (46:23.588)
Ha ha!

Alison Wilder (46:40.873)

Dr. Greg Wilder (46:43.25)
actually, depending on how much of an impact the overall harmony and timbre and texture of the music gives it. So in other words, if I create big, huge downbeats with very small filigree, we'll say, and, you know, and I do the big downbeats very slowly, that will feel lumbering and powerful and whatever. If I instead make those very fast moving bits.

Alison Wilder (46:54.815)

Dr. Greg Wilder (47:12.606)
have lots of accents that pop way out and are equal or greater intensity to the to the bar, I will it will feel much faster than that. That's a perception that we get. That's the basic idea.

Alison Wilder (47:25.412)
And this completely transcends genre or type of music. There's no, you don't even have to think about that. I can think of dozens of examples of slow music with fast on top textures from all kinds of different music.

Dr. Greg Wilder (47:31.559)

Dr. Greg Wilder (47:40.886)
And whether it's universal or not, doesn't matter to me. It certainly is true enough to have an effect on what almost anyone in today's world would find to be mood or, you know, meaning. So I'm not going to be scientific about it, but I know that this does work and it's an important

Alison Wilder (48:00.043)
true enough to have an effect. That's an interesting difference between sitting in academia and having to write papers and being, let's say, an industry and having to sell things. You go from having to be like 99.9% correct and prove it to, yeah, if you're 80% correct, that's pretty good.

Dr. Greg Wilder (48:18.99)
Well, and more to the point in academics, to me, you have to get down to some tiny, tiny little sliver of an aspect of what you really want to do and say. And you have to spend two or three years publishing on that and testing it. And so this is aside from the fact that I needed to keep my intellectual property when I was striking out with this stuff, I left university because I knew, I knew that I didn't want to go down that path. I would spend my life grinding out small details of a small thing. And I thought I could get.

Alison Wilder (48:29.875)
Yeah. Oh my god.

Dr. Greg Wilder (48:48.47)
much more of an effect by just making it work. But if I can, if I can, yeah, that was your, your trajectory full speed. Um, I want to go back to the George thing though, which is to say, w we were just talking about BPM and tempo and the idea of what that is, because you can have something as we just described at 72 beats per minute. And it will feel slow. You can have it at 72 beats per minute. It will feel extremely fast and. Or yeah.

Alison Wilder (48:51.596)
Yeah, tell me about it.

Alison Wilder (49:14.075)
Or at least active, yeah. And that is such an important difference to note if you're talking about two pieces of music.

Dr. Greg Wilder (49:21.066)
So it's worth noting, and George would go on about this a lot, like talking about what is the tempo. Tempo mattered a lot to George, the right tempo. So he would say, you know, like in, the metronome was a fairly late invention, by the way. It wasn't, you know, to have a machine that could take time consistently was something. So people, composers did not necessarily write down 72 beats per minute.

Alison Wilder (49:43.155)
Accurately, yeah.

Dr. Greg Wilder (49:49.838)
earlier, early composers. Instead, they would write Largo. They would write Andante. They would write these Italian words.

Alison Wilder (49:56.031)
So it was speeds that are related to something in life. So walking speed. Yeah.

Dr. Greg Wilder (50:03.774)
Yeah. Or something in nature or people would get a feeling of what Andante meant or what Presto meant. But George would point out, and of course this is true, Presto or Andante in Handel is not the same as Mozart and it's not the same in Beethoven. These things change over time. They change on the musical context. There's no defined tempo. When people say that classical musicians are just

repeating the score. Anyway, that's just absolute nonsense. There's so much that goes into doing that. And why these why hundreds of people record the Beethoven sonatas because their takes on them are extremely different and unique in very important ways. But tempo is a piece of this. And in fact, I mean, I always love this one, the Glenn Gould. He recorded the Goldberg variations twice and.

The first recording was what 40 minutes or 35 minutes and the second recording is an hour That's insane the difference in temp now There's reasons he's taking repeats and things that but it's also his temp tempo is different

Alison Wilder (51:09.351)
He had a really insane relationship with tempo, I would say. I'll never forget the first time, the first time I heard Glenn Gould play, I hit play on him playing the Mozart sonatas, and I just burst out laughing because they were obsessively fast, insane sounding, and I'd never heard anything like that.

Dr. Greg Wilder (51:13.382)
Yep, very unique.

Dr. Greg Wilder (51:28.204)

Yeah, you think your CD player at the time was broken or something like, you know, or something's wrong, it's playing back too fast. But tempo matters so much. And so my point is that in this metaphysical conversation, in talking about that with other composers and as a student composer, it has everything to do with how that feels and how you're going to express that feeling at whatever the BPM marking is.

Alison Wilder (51:35.94)

Dr. Greg Wilder (52:01.47)
How does it, what's the emotional impact? What's the meaning that you're conveying? And so I took that.

Alison Wilder (52:08.967)
And this conversation about this one very specific aspect of music, as you said, this conversation can be had over and over again because music has, I would say 10 or 15 aspects that you would have to look at in these same ways. And then not only would you have to look at them, do you have to look at them separately. Sorry, I keep hitting my stupid mic.

Dr. Greg Wilder (52:14.286)

Alison Wilder (52:36.539)
not only do you have to look at them separately, you have to look at how they interplay, so how they interact with each other. And that I would say, if there's one, if there's one golden goose in terms of musical identity, that's what it is. You have to look at these patterns of tension.

and relaxation of anticipation on all these levels and how they work together to form what we consider to be a single piece of music. Yeah.

Dr. Greg Wilder (53:01.698)
Well, okay.

Dr. Greg Wilder (53:12.998)
Yep. Or momentary affect. You're exactly right that you take all of this from the GTDM in terms of blocking and chunking and segmenting bits of salient music to the NARMOR expectations to the collective Huron predictive statistical training that we have to how good or bad we feel at certain points of tension and release.

to the idea of what is rhythm anyway, and how do we perceive it, and what effect does it really have? I mean, on and on and on. You have to layer all those, weight them differently, and that begins to give a machine a sense, if you can codify those things, it begins to give a machine a way to identify and separate and categorize and search on mood and effect.

of music, but it's a very complicated cognitive approach that has to be taken.

Alison Wilder (54:05.809)

Alison Wilder (54:10.255)
Right, and the machine does not have that history of statistical learning that we have. Yeah, you'd have to teach it, right? And you'd have to teach it some of these things that we're talking about, which is kind of the work that we did with Clio was about that. I read recently, I saw that there's a lab that's,

Dr. Greg Wilder (54:19.946)
You could probably get it with a trained model.

Dr. Greg Wilder (54:34.606)

Alison Wilder (54:39.199)
that's working on recreating, well, there are lots of people doing this. It's pretty intense work on using AI models to recreate patterns that are happening in the brain. So trying to figure out what people are thinking by somehow placing electrodes in different places in different ways, measuring the activity of the brain and then recreating, yeah, observing the brain.

Dr. Greg Wilder (54:55.715)

Dr. Greg Wilder (55:04.098)
But observing the brain, observing the brain knowing what? Yeah.

Alison Wilder (55:08.671)
physically, with an electrical interface, and then using an AI to look at those signals that it's producing in different people and trying to reproduce what that person is thinking. And this is obviously terrifying in certain ways, and also, like most terrifying things, also extremely exciting, the idea that we could read our own minds. But I saw recently.

Dr. Greg Wilder (55:20.382)
in different people.

Dr. Greg Wilder (55:24.846)

Dr. Greg Wilder (55:33.548)

Alison Wilder (55:37.927)
you know, someone was able to do this and reproduce, someone was listening to Pink Floyd's Another Brick in the Wall. And they were able to produce a very, a very fuzzy version, but hearable version of that. And if you haven't heard that, go look it up. It's really cool. And it really reminded me of, you know, what I always imagined Cleo was kind of

It's like the reverse process. We were trying to take these very clear to us musical recordings and the machine was getting this, this version of it that was real and was close to, it was there, but it wasn't crystal clear, right? And this had to do with a lot of technical issues with doing that, but yeah, I think I.

Dr. Greg Wilder (56:08.206)
That's right.

Dr. Greg Wilder (56:32.218)
In my work after Clio in the 10 years since then, a lot of what I did was focused on this kind of thing where I would build a model from input, take the model, separate it from the input, and then have to reconstruct it, specifically to see how accurate was that model, how fuzzy is that model, how do I want it to be represented really, because there's a lot going on there to do that.

But anyway, that's exactly the kind of experimentation I was working on or have been working on.

Alison Wilder (57:05.039)
Yeah, yeah, so that translation between, you know, between our cognitive processes and the real world and then throw in, you know, throw in a computer to that mix and you get a very, very complex and difficult problem.

Dr. Greg Wilder (57:24.058)
But imagine the creative possibilities, right? So for creativity to come. So let's say I take a piece of music that I like, I love the shape of it, all the nuances, and I make an accurate model of it. But now what I wanna do is I wanna keep that, some aspects of the core model, the tension and release, the expectation, the realization of those expectations. I wanna keep those concepts of that model, but I want totally different notes in play, totally different music, totally different instrumentation and rhythm.

Alison Wilder (57:50.151)

Dr. Greg Wilder (57:52.982)
Totally different tempo, maybe. But so you see what I mean? I can take the core identity. This is what we're talking about here. The core identity of this piece and transplant it into, like the soul of the piece, transplant it into a new piece with a different surface.

Alison Wilder (57:58.219)

Alison Wilder (58:09.771)
That's so interesting. I mean, the surface, it's funny that you say that. I've been actually doing a songwriting exercise for myself recently. That's exactly that. Just personally, where I take a song that I like and I try to make a version of it that uses the structural framework of it on that abstract level, not, oh, these chords, no.

Dr. Greg Wilder (58:19.808)

Dr. Greg Wilder (58:37.134)
Mm-hmm. Yep.

Alison Wilder (58:39.351)
It has this pattern of tension release. The sections are this long. Yeah.

Dr. Greg Wilder (58:42.954)
It's in it because you can limit yourself in interesting ways. Well, you can say this phrase is eight bars. So I'm going to keep my phrase to be eight bars. It has this pattern of tension and release. I'm going to make my own harmony have that pattern. The lyrics go here to there. I'm going to do something that else that's here to there, right? You can, is that what you mean? You're mapping these. Yeah.

Alison Wilder (58:59.383)
Yeah, yeah, right, exactly. And so my question for myself is, I was curious about this and it's related to what you just said. Is it all the musical details that make us love it so much? I'm taking like classic things like landslide. You don't know. Okay, it's a really famous song by Stevie Nicks and Fleetwood Mac did it. It's like a seminal, beautiful love song.

Dr. Greg Wilder (59:15.967)
I don't know what that is. Landslide.

Dr. Greg Wilder (59:24.311)

Dr. Greg Wilder (59:28.27)

Alison Wilder (59:28.863)
And is it the surface? Is it, you know, one thing we haven't talked about in terms of identity, and I think any listener will be like, why aren't they talking about this? Is a person's voice, right? So if we're talking about music with voice, that's the first thing anyone would think about. Well, I identify music by the person singing it, which is huge, obviously, that's true. You know, we're trained, you know, talk about.

Something our brains are statistically trained on is recognizing and understanding the human voice. That's like the most important.

Dr. Greg Wilder (01:00:04.142)
Well, okay. And what's, it's also what's a value. Case in point, Katy Perry, I think last week, just sold her entire music catalog to an investment firm. You know this? For $350 million. Now, the value, is the value in the keyboardist? Because they were, it's all of her like five, six albums or something. Is it in the keyboardist? Is it in the bass player? Is it in the drummer? Where's the value? The value is in her voice, her.

Alison Wilder (01:00:14.155)

Alison Wilder (01:00:29.576)

Alison Wilder (01:00:32.683)
Her brand, basically. Yeah, because her voice becomes a proxy for something larger and cultural. So yeah, I guess that's an important thing to point out that, yes, the human voice, yes, yes. But in current contemporary music, that most people are putting in their earbuds.

Dr. Greg Wilder (01:00:33.662)
Her brand, her persona exactly and all. That's it.

Alison Wilder (01:00:57.911)
it's a lot more complicated than that. And there are business interests that are dictating, you know, a lot of what's happening. And those things are not what we're talking about here. When we talk about musical identity, and it's really hard because that's been such a huge part of our lives and our listening, you know, record labels have determined a lot about how we listen to music, for sure. And so...

It's hard to parse, you know, what is cultural and what is musical at this point, at least for me. I have to think really hard about it when I try to do it and be sure that I'm not letting non-musical things bleed into my musical, purely musical thinking. Not that there's anything wrong with talking about cultural things, but if you, yeah.

Dr. Greg Wilder (01:01:41.258)
I think it would be, I don't know if this is a weird thing to say, it feels weird, but it's this, if could you get an investment firm to pay $350 million for Beethoven? Right. If that were something you could own, and I don't think you could, because what value does that have? Katy Perry will have a value for the next 10, 20 years. But her value today, right now.

Alison Wilder (01:01:53.519)
All, like all Beethoven? That's a great question.

Alison Wilder (01:02:06.763)

Dr. Greg Wilder (01:02:10.762)
monetarily speaking, is far higher than Beethoven. So clearly, there are extra musical things going on there. I mean, that's exactly right. She became a brand. She has there's a lot of cachet there. So there's a way that they're going to recoup that value and more. And they're positioned to do that. So they're going to do it as a business. But as you say, very important point. We are not talking about that kind of value or identity. We're talking about the core identity of the music. And why is that more interesting?

to me anyway, because it's not the immediate stage of where AI is going to be, where they're cloning people's voices. That's pretty simple, actually. It's not simple, but it's a pretty basic way of grasping at somebody's identity. Much more important is conceptually simple. But what's much more interesting is someone's identity and their body of work, their

Alison Wilder (01:02:55.611)
It's conceptually simple. It's technically hard, conceptually simple, yeah.

Dr. Greg Wilder (01:03:05.89)
the way they play, the way they compose, the meaning that they produce, the effect that they can have. This goes for film composers and all wide range of anybody making music that has an impact socially. How does that?

Alison Wilder (01:03:19.715)
Yeah, so if I had to wrap it with a bow, it's that the kind of musical identity that we're trying to understand and talk about today is the kind that makes you know when you just hear something out of the blue, not from the voice, but just from listening to the music, you know who made it. I know that this is

Dr. Greg Wilder (01:03:24.191)

Alison Wilder (01:03:48.359)
You know, I can listen to six seconds of it and I know this is Radiohead, for example. Or, you know, I'm... Yeah, we don't have to hear the whining to know. Um... Ha ha.

Dr. Greg Wilder (01:03:53.426)
Even if you took out the voice, okay, the whining of, yeah. If you took out like Janet Jackson, right? You take out Janet Jackson's voice and you play some unknown track, could you tell that it was produced by Jimmy Jam and Terry Lewis? I bet you could, right? Or...

Alison Wilder (01:04:09.767)
No, but you could, yeah. Yeah, and that's an interesting point. Producers are really who's creating, composers create the identity in Western classical music. Producers are really who's creating the identity, I would say. And it's funny that the entirety of the LA structure, the major label structure, is designed to talk about the artist, right? It's all about the artist. And that means the singer, who probably had, in most cases, actually very little to do with the creation of

anything except what came out of their mouth in the mic that one time, which was then totally changed after the fact at this point in modern music making. That's how it works. So the producers are creating that identity. So can you hear a bit and tell it's Timbaland? And our favorite producers are people where the answer is yes. Yeah, I can tell it's Timbaland. And again, not just because he's, you know, kind of shouting in the background, which he's want to do, but

because he has a way of making beats that's very identifiable. And how to codify that, that's the type of identity that we're talking about. And that is hard and cool.

Dr. Greg Wilder (01:05:14.194)
And it's essential, it's essential because that has in my, my estimation, more of a cultural impact long-term than Carrie, Katy Perry's, uh, five or six albums. You know, that's a very different kind of business exchange, but a Timbaland, uh, groove is going to have an impact to so many artists for so long. Culturally, it's going to have, it's going to mean some things. It's going to be very important. And.

And the AI machines are going to pick that up and they're going to start cloning that in some way. And we need to understand.

Alison Wilder (01:05:46.963)
Well, and what's happening with LLMs as they're feeding themselves back their own creation, what's been found is that, you know, if an LLM is trained on LLM generated material, it gets worse and worse and worse. It's not good. It only is working when it's trained on human created material. And so I think that, you know, as music systems, AI systems get better,

I think you will find that as well, unless you can teach a computer to understand all of these things about creating music that humans intuitively have understood just by listening and very talented and hardworking humans have figured out how to understand it and make it, right? And so, I think it's worth noting that you can't just expect a computer to learn

to produce great music, unless great music is still being made to teach it.

Dr. Greg Wilder (01:06:52.566)
Well, and that brings us to a whole other potential podcast discussion, which is that of the process of evolution in music invention. So can you take what would be Darwinian evolution, apply it over many, many iterations to find mutations that are going to be satisfying? And could that produce a creative machine?

Alison Wilder (01:07:21.099)
Am I talking to Richard Dawkins right now? Is that, you talking about like the Dawkins style meme, the idea meme? Yeah.

Dr. Greg Wilder (01:07:23.147)

Dr. Greg Wilder (01:07:26.634)
Absolutely. The meme, yes. The idea that a meme will live through, it does live through, so an artist can create it whether they know about the past or not, and it can be injected into the system and if it takes off, and it has to, I'm writing an article about this, it's coming out soon, but if it has exposure and it can grab hold, it can have an influence and grow and become part of the, you know, part of the fabric. Just like a

just like a biological evolutionary process. There's a lot of similarities there.

Alison Wilder (01:08:00.535)
And in music, in order for an idea to be powerful like that, I think it has to have a strong, clear, and pleasant somehow, or desired identity, right? That's what actually, I guess that's how we're really, that's where the rubber meets the road with musical identity. It's does it live on? Does it influence others, right?

Does it inject itself into the musical stream of production? Yeah, and that is absolutely.

Dr. Greg Wilder (01:08:28.759)
Well said.

Dr. Greg Wilder (01:08:35.714)
There are things that have to, there are things you can't do that inside a machine. This is key. You can't just have an iterate inside the machine and decide this is the evolution because you haven't given it the test of. Ed hasn't had a chance to go out and live. These mutations haven't had a chance to go out and try their, try their luck or whatever, see how they grab an audience.

Alison Wilder (01:08:56.523)
So you're saying if, unless we can codify what makes musical identity successful in that way. So you're saying if we can do that, if we can codify this is what makes musical identity successful, it has held true over this long period of time across cultures, you know, these are what make it work. These are the things that we can define. If you can define that to a point, to a fine enough point that you can teach a computer what that means.

Dr. Greg Wilder (01:09:06.036)

Alison Wilder (01:09:24.647)
You're saying then that's when a computer can create its own musical, its own ideas with their own musical identity.

Dr. Greg Wilder (01:09:32.334)
Because it has to, yes, that's right. It has to judge. It has to judge in a way that we would, and I say we collectively, not individuals, but the way we would as a group. And that leads you to the search for universals, if there are some, or universal concepts, which is another podcast topic. But yeah, so yes, you nailed it. Yes, I believe you could create real computational creativity.

Alison Wilder (01:09:57.615)
All right.

Dr. Greg Wilder (01:10:02.892)
in that way.

Alison Wilder (01:10:05.127)
Well, let's leave it at that for today. It's a lot. This is obviously one of our absolute favorite topics. You will hear us continue to talk about identity forever if you're willing to listen. But for today, we're going to wrap it up. Thank you so much for listening. And yeah, we hope to hear from you. We hope to see you.

Dr. Greg Wilder (01:10:08.322)
a lot.

Dr. Greg Wilder (01:10:25.538)
Thank you.

Alison Wilder (01:10:30.431)
Please feel free to write us if you have questions or just want to chat or you have ideas for topics that you'd like to hear us cover. And other than that, we're gonna see you soon. Have a great day.

Dr. Greg Wilder (01:10:42.242)
Be well.