![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Entry tags:
generated music my version
My goal here with this journal, the YouTube channel, the soundcloud page, is to create a algorithm generating music using the pure data programming environment. A pure data patch that emulates a rehersal band playing an ever evolving composition. Like, Ambient drones set to fluctuate over time, audio granular fields evolving over time or like that of a sample & hold module creating an ever evolving score with the voltage notes from noise. I want to develop a patch that functions primarily like a generative rehearsal band. A pure data patch structure employing; randomization, probability, comparative probabilities, expression mathematics & even crazy fractal type musical algorithms. Mathematical concepts that hopefully lead to an algorithm that is more of a human equation. I hope to tame the chaos into a recognizable form of music. A form of music that promotes accompaniment from other musicians. Building a bridge between the computer's algorithm and the human composer & musician accompaniment. I hope to play along with these compositions & not find ourselves an electrical needle in a static glitch haystack.
For a work to be generative it should evolve over the course of it's production; rhythm, notes & chords all moving with an element of change during the course of time in the composition, never to be redundant. Any other type of music composition that finds it's self repeated for the cohesiveness to the expression, falls into whatever genre of music it has been predetermined to be through repetitive tempo & tonal choices. Reinventing a common pop song or bringing back from the dead a renown musician to play in a new musical composition using the generative process is not much of a creative endeavor in the pursuit of new music? More a historical recreation presentation & possibly a bastardization in the eyes of the deceased musician & their fans. An argument has been evoked here and will play out in my journal & elsewhere. Should we humans revel in our own creativity or give praise to artificial creativity? Will artificial creativity yield a genre all of it's own? Or will artificial creativity just continue to recycle imitations & bastardizations of human creativity? Does the computer listen to us or are we going to stop and listen to the computer? And for a much deeper discussion, are we through listening to ourselves & would rather listen to a machine???
If we are going to listen to a machine, then let it be something original in the genres of music. Let us program the machine to create new compositions, not train it to recycle the music of our culture. There are powerful companies at present spending vast amounts of money to create a technology that frequency analyzes audio & musical samples of other musicians to create recycled bastardizations of that original all to avoid copy-write laws. That is pretty sad & certainly not really becoming of a global company.
Pure data is a fascinating musical programming environment that has taken me a few years to incorporate into my musical toolbox, I am almost at a point with the application where I can generate a musical composition, emulating the sounds made by a number of musicians playing in harmony with each other. Through it's algorithm construction generating a new an evolving piece of music that allows me to join in and play along. Sure we have been able to do this with drum machines, sequencers & digital workstations. But those compositions are built like lEGOs and often very repetitive in composition. What I'm describing is more like a rehearsal band in a constant state of improv, yet through the chaos developing a song that stays on common ground and doesn't wander too far off into the unknown.
Consider the compositions of Edgard Varese as a early form of a generative evolving composition.
These early pioneers of alternative compositions in their day were definitely tuned into evolving audio. For a work to be generative it should evolve over the course of it's production; rhythm, notes & chords all moving with an element of change during the course of time in the composition, never to be redundant. Any other type of music composition that finds it's self repeated for the cohesiveness to the expression, falls into whatever genre of music it has been predetermined to be through repetitive tempo & tonal choices. Reinventing a common pop song or bringing back from the dead a renown musician to play in a new musical composition using the generative process is not much of a creative endeavor in the pursuit of new music? As I have stated earlier, more a historical recreation presentation & possibly a bastardization in the eyes of the deceased musician. The philosophical argument has been evoked and will play out in my journal & elsewhere. Should we humans revel in our own creativity or give praise to artificial creativity? Will artificial creativity yield a genre all of it's own? Or will artificial creativity just continue to recycle imitations & bastardizations of human creativity?
Many forms of evolving music like ambient music are generative, Improv evolves, solo instrumentalist can take us on evolving journeys within any song. My goals in this endeavor will be to create an evolving form of music that performs with a more humanistic feel than the common glitch & manipulated frequency compositions.
This genre of music is breaking ground and should have some definitions that identifies it as a generated evolving music composition. As you search through soundcloud using the available tags that artists are assigning their music in this portal, you quickly realize that a vast majority of the artists on soundcloud tagging their music as generative couldn't be farther from the the generative genre truth. It's one of the first problems I have experienced with soundcloud, " How to get in touch with a specific genre if all these soundcloud accounts are so inclined to tagging their music incorrectly to a genre...when obviously is not a composition in this genre?". Generative music doesn't create pop songs, but yes you can generate a pop song through the use of matrices picking apart the audio file of an existing pop song. There are some excellent degenerative pure data compositions on the internet that come interestingly close to creating pop songs. An attempt to create a pop music track would require editing & reediting a generated midi score out of it's generating evolving performance. Creating a generated composition in a specific genre would be a roll of the in-coherent dice every time. But let us make the point about soundcloud. Because of the way people are tagging their music, incorrectly, what we find is that soundcloud is like a record store where every genre in the store is getting categorized incorrectly. Not necessarily soundcloud's fault, but because soundcloud allows for the account owner to tag their uploaded track as whatever, it creates this vast amount of music that is just smeared across all of the other genres. It's as if when a word like "generative" becomes part of the national conversation, we have a legion of producers moving their music from one category to the next in the hopes of more exposure. It creates a muddy audio mess. So when on soundcloud, and you go searching for generative music, get ready to have to listen to everything but generative music. This only creates confused listeners? Asking themselves is this a generative song? Coz it sure sounds like country pop to me. Or is it that, who cares? It's only music right? I don't remember ever going to Tower Records & the owner telling me that I can find all the classical music in the rock, country & disco sections.
It is easy to envision two groups of artists using generative processes to create music. One, There will be those who will use AI tech to generate more songs that will sound like existing songs. Like what Google's musicLM or Lyria promotes. They will use advanced sampling technology to analyze old songs, frequency analyze the music it consumes into a matrix, cataloging those scores into matrices, recalled when tasked to create new ones. This process in the long run will suffer the same fate as those who sampled others to colorize their own music. In that if an original can be drawn from the copy, credence will have to be given to the creator of the original, through everything that copy does. This is sort of dead end when it comes to creating new music. Two, there will be musicians who will delve into the generative process to create something new. Construct the chaos into a audible definitions to be understood by all. Pure data can be using as a powerful composition tool in music. Think of it this way, if midi in 1983 was midi 1.0. Pure data midi is definitely midi 2.0 . Midi gets the music out of the PC. A midi composition creates an accountable path in the composition new music. As valid as any written musical score.
A midi artist, creating a generated composition can secure the components of their patch; performance and tables of sounds, unique samples, the programming of the conditions & randomness. Performed many times over in a studio setting until the single take recording is perfected. This is the process that I endure. When I perform a live generated midi composition the performance will is never the same twice. To found a generative genre of music must move this evolving element of the composition into the foreground. Like a singer is brought to the front of the stage. In my music it's the angle and the process that got me here and it will be that evolving characteristic that gets the spotlight.
Being able to perform the pure data patch is just as important as it's recording. A pure data patch, connected to outboard gear or not, should have a certain level of live consistency. In the end we are talking about creating songs. I don't think of myself as an audio artist because I have figured out how to make my hardware do something, I think of myself as an audio artist because I attempt to get out of the hardware something. Constructing songs is important, therefor composition is important. And in this generating genre the evolving composition is important. It defines it. Without it...it's just generated. : ( Why be generated, when you can be generating! : )
A huge injection of improvise would be another way of perceiving a generated score. If I was to put this attribute or tag on a 4 piece human band, I would come away from a listen of that band stating that the band never stopped jamming, never returned to a point from which they started. In a perpetual state of audio composition evolution. So why label your soundcloud track generative if it is a repetitive routine or a recycled production of an existing song of some genre? It would be like a country western performer claiming that they were a industrial death metal artist. Doing so only obscures the truth from the listener and in the end reveals itself as a fake.
Can style be created in generative music? That is one of my key directives for producing this journal & uploading my tracks to soundcloud. During this journey I will try to point out the areas in my pure data compositions that are dictating change & how I produced that particular instrument. I also want to point out where in my composition that you can add levers, sliders & knobs in the patch where you can change the generative process in real time. Key & time signature will help to create style, though sometimes my early patches have caused the song to leave key & time signature. If the programming for key & time signature changes are there, then the random equations following will continue to determine new outcomes. Polymeter; tuplets & quintuplets, will add more to the pure data composition's diversity in the style. So where the midi notes go, only the programmer knows? The programmer of a pure data code or structure only has an idea of what they want, for each live performance of a generated composition yields a different result. Even in the way you describe a generative composition is a departure from the way musician's have described their music in the past. A generative song is very different from previous forms of music, and yet is still just another genre.
At this point as a generative music artist, I am creating three final performances that have their foundation in Pd & I will label them as such in their track descriptions on soundcloud. The first is RAW, where I make no attempt to edit the outbound midi performance of the generation. Just let pure data produce the notes & make no attempt to harness them. DIRECTED, where I make no attempt to edit the outbound performance, but put in place conditional programming or live manipulation using controllers. EDITED, where I take a large mass of generated midi score and sculpt a song out of it. It's the same song, it's just had it's measures rearranged, notes cut & pasted...so on. Each is very distinguishable from the others. If there is a last form, ACCOMPANIED, Where either I play along with the generated score on other instruments or pure data accompanies me. Of my tracks, strands together & The Coil, currently uploaded to soundcloud are examples of me accompanying a generative composition.
These forms of a generated midi score; RAW, DIRECTED, EDITED, ACCOMPANIED, will be defined in my generating compositions of the generative genre.
For a work to be generative it should evolve over the course of it's production; rhythm, notes & chords all moving with an element of change during the course of time in the composition, never to be redundant. Any other type of music composition that finds it's self repeated for the cohesiveness to the expression, falls into whatever genre of music it has been predetermined to be through repetitive tempo & tonal choices. Reinventing a common pop song or bringing back from the dead a renown musician to play in a new musical composition using the generative process is not much of a creative endeavor in the pursuit of new music? More a historical recreation presentation & possibly a bastardization in the eyes of the deceased musician & their fans. An argument has been evoked here and will play out in my journal & elsewhere. Should we humans revel in our own creativity or give praise to artificial creativity? Will artificial creativity yield a genre all of it's own? Or will artificial creativity just continue to recycle imitations & bastardizations of human creativity? Does the computer listen to us or are we going to stop and listen to the computer? And for a much deeper discussion, are we through listening to ourselves & would rather listen to a machine???
If we are going to listen to a machine, then let it be something original in the genres of music. Let us program the machine to create new compositions, not train it to recycle the music of our culture. There are powerful companies at present spending vast amounts of money to create a technology that frequency analyzes audio & musical samples of other musicians to create recycled bastardizations of that original all to avoid copy-write laws. That is pretty sad & certainly not really becoming of a global company.
Pure data is a fascinating musical programming environment that has taken me a few years to incorporate into my musical toolbox, I am almost at a point with the application where I can generate a musical composition, emulating the sounds made by a number of musicians playing in harmony with each other. Through it's algorithm construction generating a new an evolving piece of music that allows me to join in and play along. Sure we have been able to do this with drum machines, sequencers & digital workstations. But those compositions are built like lEGOs and often very repetitive in composition. What I'm describing is more like a rehearsal band in a constant state of improv, yet through the chaos developing a song that stays on common ground and doesn't wander too far off into the unknown.
Consider the compositions of Edgard Varese as a early form of a generative evolving composition.
These early pioneers of alternative compositions in their day were definitely tuned into evolving audio. For a work to be generative it should evolve over the course of it's production; rhythm, notes & chords all moving with an element of change during the course of time in the composition, never to be redundant. Any other type of music composition that finds it's self repeated for the cohesiveness to the expression, falls into whatever genre of music it has been predetermined to be through repetitive tempo & tonal choices. Reinventing a common pop song or bringing back from the dead a renown musician to play in a new musical composition using the generative process is not much of a creative endeavor in the pursuit of new music? As I have stated earlier, more a historical recreation presentation & possibly a bastardization in the eyes of the deceased musician. The philosophical argument has been evoked and will play out in my journal & elsewhere. Should we humans revel in our own creativity or give praise to artificial creativity? Will artificial creativity yield a genre all of it's own? Or will artificial creativity just continue to recycle imitations & bastardizations of human creativity?
Many forms of evolving music like ambient music are generative, Improv evolves, solo instrumentalist can take us on evolving journeys within any song. My goals in this endeavor will be to create an evolving form of music that performs with a more humanistic feel than the common glitch & manipulated frequency compositions.
This genre of music is breaking ground and should have some definitions that identifies it as a generated evolving music composition. As you search through soundcloud using the available tags that artists are assigning their music in this portal, you quickly realize that a vast majority of the artists on soundcloud tagging their music as generative couldn't be farther from the the generative genre truth. It's one of the first problems I have experienced with soundcloud, " How to get in touch with a specific genre if all these soundcloud accounts are so inclined to tagging their music incorrectly to a genre...when obviously is not a composition in this genre?". Generative music doesn't create pop songs, but yes you can generate a pop song through the use of matrices picking apart the audio file of an existing pop song. There are some excellent degenerative pure data compositions on the internet that come interestingly close to creating pop songs. An attempt to create a pop music track would require editing & reediting a generated midi score out of it's generating evolving performance. Creating a generated composition in a specific genre would be a roll of the in-coherent dice every time. But let us make the point about soundcloud. Because of the way people are tagging their music, incorrectly, what we find is that soundcloud is like a record store where every genre in the store is getting categorized incorrectly. Not necessarily soundcloud's fault, but because soundcloud allows for the account owner to tag their uploaded track as whatever, it creates this vast amount of music that is just smeared across all of the other genres. It's as if when a word like "generative" becomes part of the national conversation, we have a legion of producers moving their music from one category to the next in the hopes of more exposure. It creates a muddy audio mess. So when on soundcloud, and you go searching for generative music, get ready to have to listen to everything but generative music. This only creates confused listeners? Asking themselves is this a generative song? Coz it sure sounds like country pop to me. Or is it that, who cares? It's only music right? I don't remember ever going to Tower Records & the owner telling me that I can find all the classical music in the rock, country & disco sections.
It is easy to envision two groups of artists using generative processes to create music. One, There will be those who will use AI tech to generate more songs that will sound like existing songs. Like what Google's musicLM or Lyria promotes. They will use advanced sampling technology to analyze old songs, frequency analyze the music it consumes into a matrix, cataloging those scores into matrices, recalled when tasked to create new ones. This process in the long run will suffer the same fate as those who sampled others to colorize their own music. In that if an original can be drawn from the copy, credence will have to be given to the creator of the original, through everything that copy does. This is sort of dead end when it comes to creating new music. Two, there will be musicians who will delve into the generative process to create something new. Construct the chaos into a audible definitions to be understood by all. Pure data can be using as a powerful composition tool in music. Think of it this way, if midi in 1983 was midi 1.0. Pure data midi is definitely midi 2.0 . Midi gets the music out of the PC. A midi composition creates an accountable path in the composition new music. As valid as any written musical score.
A midi artist, creating a generated composition can secure the components of their patch; performance and tables of sounds, unique samples, the programming of the conditions & randomness. Performed many times over in a studio setting until the single take recording is perfected. This is the process that I endure. When I perform a live generated midi composition the performance will is never the same twice. To found a generative genre of music must move this evolving element of the composition into the foreground. Like a singer is brought to the front of the stage. In my music it's the angle and the process that got me here and it will be that evolving characteristic that gets the spotlight.
Being able to perform the pure data patch is just as important as it's recording. A pure data patch, connected to outboard gear or not, should have a certain level of live consistency. In the end we are talking about creating songs. I don't think of myself as an audio artist because I have figured out how to make my hardware do something, I think of myself as an audio artist because I attempt to get out of the hardware something. Constructing songs is important, therefor composition is important. And in this generating genre the evolving composition is important. It defines it. Without it...it's just generated. : ( Why be generated, when you can be generating! : )
A huge injection of improvise would be another way of perceiving a generated score. If I was to put this attribute or tag on a 4 piece human band, I would come away from a listen of that band stating that the band never stopped jamming, never returned to a point from which they started. In a perpetual state of audio composition evolution. So why label your soundcloud track generative if it is a repetitive routine or a recycled production of an existing song of some genre? It would be like a country western performer claiming that they were a industrial death metal artist. Doing so only obscures the truth from the listener and in the end reveals itself as a fake.
Can style be created in generative music? That is one of my key directives for producing this journal & uploading my tracks to soundcloud. During this journey I will try to point out the areas in my pure data compositions that are dictating change & how I produced that particular instrument. I also want to point out where in my composition that you can add levers, sliders & knobs in the patch where you can change the generative process in real time. Key & time signature will help to create style, though sometimes my early patches have caused the song to leave key & time signature. If the programming for key & time signature changes are there, then the random equations following will continue to determine new outcomes. Polymeter; tuplets & quintuplets, will add more to the pure data composition's diversity in the style. So where the midi notes go, only the programmer knows? The programmer of a pure data code or structure only has an idea of what they want, for each live performance of a generated composition yields a different result. Even in the way you describe a generative composition is a departure from the way musician's have described their music in the past. A generative song is very different from previous forms of music, and yet is still just another genre.
At this point as a generative music artist, I am creating three final performances that have their foundation in Pd & I will label them as such in their track descriptions on soundcloud. The first is RAW, where I make no attempt to edit the outbound midi performance of the generation. Just let pure data produce the notes & make no attempt to harness them. DIRECTED, where I make no attempt to edit the outbound performance, but put in place conditional programming or live manipulation using controllers. EDITED, where I take a large mass of generated midi score and sculpt a song out of it. It's the same song, it's just had it's measures rearranged, notes cut & pasted...so on. Each is very distinguishable from the others. If there is a last form, ACCOMPANIED, Where either I play along with the generated score on other instruments or pure data accompanies me. Of my tracks, strands together & The Coil, currently uploaded to soundcloud are examples of me accompanying a generative composition.
These forms of a generated midi score; RAW, DIRECTED, EDITED, ACCOMPANIED, will be defined in my generating compositions of the generative genre.