jnWake

Your social media powers are too strong haha. EDIT: Lmao I checked and one of the comments was even originally in Spanish. Truly a tamer of lions.

This one got posted fast haha, always nice when that happens! I'm very proud with how it ended up, collaborating with Emunator worked out great and it was super fun. The writeup on the mixpost page has this line as a quote: "Hehe, there is always room for new things, even for the most "overcovered" themes. The 2 tracks go surprisingly well together, don't they?"... I'm fairly sure I never said that haha, was that Emunator?

Wow at the album you linked in the description Larry, it's ridiculous. I can definitely understand the comparison between that and this remix haha.

I mean, OP doesn't mention "basic prompts" or anything of the sort, it just mentions prompts. The "simple" part was by me 😋 There's really 2 points in the OP, one is about the ethics of how the models were trained regarding artist consent and copyright. I'm not sure if OCR's stance would be the same if the models were "ethically" trained or if, for example, a user trained their own model only on songs they made or something like that. There's a second point about the "interpretation" aspect of AI generated music, the "human touch". I think we simply disagree here, I don't think current tools provide much "human touch". Even if Suno allows for more than "a simple prompt", the "human part" is still a low percentage of the finished creation. Of course this could change with time and how the technology evolves but it's my stance on what I've seen of current technology. In the end, it's up to the community to decide how it approaches these things. I think it's fair if OCR decides it wants to focus on music that is created on a larger percentage by the "human touch" and, of course, policies can change over time as technology evolves.

I mean, the rule discussed in the OP is pretty specific, regarding websites like Udio, which are quite dubious (ethically) in how they were trained. OP is not banning all uses of AI in music and points it out clearly. This is not about being technology myopic and saying "AI is the devil" or something silly like that. This isn't about "you can't use tools in music" or anything similar either. For example, I begin all my mastering from Ozone's "smart master" function, which uses AI. I don't see the point in this honestly. Sure, a big percentage of stuff we use in music is, to some degree, a black box (like, I have no idea how an EQ actually does what it does), but the difference in how much control I have between those plugins and (today's) genAI tools is night and day. Although I don't know exactly how my VSTs generate the notes that they generate, I tell them what note to play and I usually have control over a decent amount of the sound properties. With genAI you don't have control beyond a prompt and stuff like modifying an output is basically impossible (with today's tools at least). As for more complex prompts like what you're saying... I dunno, it seems like a technology so completely different to what the OP is discussing that I don't think bringing up it as an hypothetical makes much sense really.

The difference is that when an artist uses paints that Da Vinci used, they're deliberately choosing how they do it. Training and then using an AI is a complete black box where you don't control the output at all beyond a simple prompt.

Echoing the question about the playlist actually existing... In case it does, I can add a few of mine:

Nice seeing this posted. I agree with the judges comments about the timings on the intro, I had a hard time adjusting the shuffles to a degree I was 100% satisfied. I kinda just left what satisfied me the most haha. I don't like to complain about judge comments too much, but this one left me so confused: Like... what?! The track even has a guitar solo near the outro. It must be the weirdest comment I've gotten from an OCR judge. It's a live solo played on the keys, I don't see how that's anything weird or unusual at all.

There’s a mistake in the writeup post, what I said in the submission was: “For better results I recruited Hakstok to play bass and Lucas to play brass. Lucas then, in turn, contacted Gamer of the Winds for the woodwinds and that ended up forming a pretty neat team!” Somehow it got copypasted wrong in the writeup!

I started the arrangement with a jazz fusion style on my mind, but prog rock started winning as I added more sections. I’ll take a look at the piano tone the judges mentioned. I’m using NI’s “The Giant” which has a variety of possible sounds, but I’m intentionally tweaking it for an aggressive and bright sound. I probably overdid it haha.

Featured again hehe, nice! I see some judges weren’t that fond of the synths but what can I say, I love ‘em. Jazz-ish stuff with synths is dope in my book.

Nice to be back on the front page after a while! Thanks to my collabers for their great performances on this. For those who read the write up, I strongly recommend the Frost* original song that inspired this remix, it’s phenomenal... as is Frost*’s music in general.

I actually do it a little different from what's been shared here. I tend to start by trying to figure out the bass (the root notes at least) and from there get the chord progression and key of the song. In my opinion, you get a lot of useful information from a song from just this. Melody is important of course, but a ton of VGM songs tend to play around the chord notes so, once you know the chords, the melody comes easy. When it comes to actually writing a remix, I tend to learn the song and then just sit down and improvise. It helps if you have a predefined idea, like, "make a metal cover", but usually just playing around with the melodies and chords is enough to trigger some inspiration.

I've uploaded 2 more tracks since my original post, hope you like 'em: #6. Castle - Pre-Boss #7. Castle - Boss Any feedback is appreciated!

This one: https://ocremix.org/remix/OCR04084