Latest Track
Selections from our first album:
A Word from the Creator of the Glass Cats
Concern about reviews or criticism of this music would have to presume that anyone will actually listen to it, of course; and that presumption is highly dubious—primarily because the probability of anyone finding it in a brave new world of independent online musical artists would be like finding a particular grain of sand on a Miami beach. But, somehow, you found it; and I sincerely appreciate anyone who takes the time to hear what I have labored with some passion and purpose to create.
I harbor no delusions of grandeur, nor do I necessarily favor a feline fantasy world over cold reality. I created the Glass Cats merely as a “fun” vehicle through which to present my music. I honestly have no desire to be “out front” anyway, so I will happily leave that to Liam and Amara. I just find the act of creation immeasurably satisfying, and when that effort occasionally leads to something “worthwhile,” the feeling is nothing short of sublime. I do this for the adrenaline rush and the sense of purpose it gives me. I do it simply because I love doing it.
Now let’s add a little controversy: I freely acknowledge that I have used that dreaded Artificial Intelligence (“AI”) in the creation of each of the final productions presented here. However, each and every piece of music presented by the Glass Cats was and is wholly written by myself—both music and lyrics. My workflow involves first making the recording of an entire production of each composition in digital multitrack, using virtual instruments (“VSTs”) and a MIDI keyboard workstation that I play myself. Ideas are fully developed into final compositions by a human being—myself. In the case of Ukraine Lullaby, for instance, the working or demo recording consisted of more than 48 tracks of instrumentation, percussion and vocals. Then each piece is mixed and mastered in the traditional way. But I will also freely acknowledge that there is a world of studio musicians who can play piano far more beautifully than I can; even more so, violin or Irish harp, for that matter. Even using state-of-the-art VSTs (and there are many phenomenal virtual instruments that can be played in real time on a keyboard controller, using key switches and modulation wheels for realistic articulations), but I have always felt that my compositions were being limited by my ability to play them as well as they might deserve.
Stage two of my workflow therefore involves uploading the demo production of a finished piece of music into an AI platform to be “covered.” This is essentially the same process a songwriter or composer follows when he brings a group of studio musicians together to play—or “cover”—the material he has written. He gives them the music and instructions on how he would like it played. Using AI, the songwriter creates specific “prompts” to detail his vision for the final production. Perhaps the strings are to play pizzicato in a section… maybe the electric guitar should punctuate the lead I played in the demo with pinch harmonics… perhaps the entire production should be colored with tape saturation to create a surreal or other-worldly sound. Artificial Intelligence essentially gives a composer access to a top-flight roster of musicians in a state-of-the-art recording studio. These are things that the vast majority of independent artists could never afford. And without it, they are forced to relegate their music to inferior productions that no one will ever hear.
Stage three of my workflow involves separating the segments of the AI production into tracks or “stems,” then importing them back into my digital multitrack work station, adding further instrumentation as I deem desirable, then mixing and mastering the final production from my original multitrack work station or “DAW” into its final stereo form.
I know there are “purists” who insist that AI is useful only to those who are incapable of good composition, and I will certainly acknowledge that AI can be used to compose material with very little human input. I don’t use it that way, and I never will. My compositions are complete before they are ever uploaded to an AI platform. I use Artificial Intelligence in virtually the same way a songwriter collaborates with a band of great musicians. I ask the drummer to do specific things, then I let him embellish the production with his own touches. I tell the lead guitarist what the hook is, then I let him create harmonies in multitrack. The major difference is that these top-flight musicians are at my complete disposal 24/7, and I don’t have to rent costly studio production time for them. Otherwise, I see very little difference.
When multitrack recording became widely used in the 1960s and 1970s, there were those purists who claimed that “real musicians” must play entire pieces of music live, and multitrack was just a gimmick used by untrained artists to make themselves sound better. There was some truth to that, of course. But that criticism did not stop some of the most talented “untrained” artists from pushing the creative envelope that multitrack provided—especially in the genre known as “progressive rock”—until even the purists had to embrace it as a production tool. I see very little difference, when AI is used as it has been used to produce these Glass Cats recordings.
I will gladly leave the purists to their endless hours of practice time and their music theory —which is really only an older composition tool, after all. Theirs can be a sad, rigid, exclusionary reality. I would rather be free to create, and to present the things I create in a light most favorable to them. I won’t ever have AI create a piece of music for me. That would be no fun at all. But I do allow AI some liberties in the way it interprets and plays the finished pieces I upload, just as I would allow talented studio musicians to interpret a song I presented to them for recording. I see very little difference, when used in this way.
But I wanted to be honest and completely transparent in acknowledging my use of Artificial Intelligence in the final production of the Glass Cats material because it is a tool that I enthusiastically embrace. Whether others admit it or not, most traditional recordings now use AI in one form or another in their productions—whether through virtual instruments or when mastering a final mixdown of a multitrack recording. Technology should never substitute for human creativity, but it can certainly be used as a tool to enhance it. And for that I offer no apologies.
I sincerely hope you enjoy these compositions and productions as much as I enjoy making them. There will be many more to come. In the meantime, the Glass Cats and I send our best wishes for your inspiration… or at least your amusement.
Eric J. Dorer