Liontamer ⚖️ Posted Sunday at 06:37 PM Posted Sunday at 06:37 PM (edited) Artist Name: Craig Jackson Chains of heroism and hidden sorrow collide in this post-grunge take on Zelda’s classic overworld theme. Crunchy guitars and driving drums channel the weight of unspoken struggles, transforming Hyrule’s adventure into an emotionally raw release. Games & Sources The Legend of Zelda (A Link To The Past) - Overworld Edited 2 hours ago by pixelseph
Liontamer ⚖️ Posted Sunday at 06:41 PM Author Posted Sunday at 06:41 PM FYI, the submitter hasn't included linked YouTubes for the source tunes, so I'm providing those. Most of their links are to SoundCloud streaming links rather than downloadable files, FYI. They've submitted way beyond what we allow in such a short timeframe, but the panel's been doing well and it almost caught up, so we'll go through these. This sounds like AI, so including some conversation we had in the submission area with the artist for reference. Asked whether this was generative AI, the answer was: "So this is a yes and no answer. I wrote the lyrics, played the guitars and bass. I added in the drums using garageband, and used AI for the vocals. So not generative AI, with Ace Studio for the vocals yes. Much like all of my other submissions, I play as much as I can with guitar, bass, piano, mandolin or violin and let garageband fill out the rest of what I can't play, do not own, as well as Ace take care of singing for me."
Liontamer ⚖️ Posted Sunday at 09:20 PM Author Posted Sunday at 09:20 PM The track was 4:36-long, so I needed to hear the VGM used for at least 138 seconds to consider the source material dominant :00-:34, 1:45-1:57, 4:20-4:33 = 49 seconds or 17.75% source usage The lack of direct source usage & substantive arrangement is an automatic dealbreaker for me. NO The volume's pretty low; not sure why it's limited like that, it could be bumped up. Opens with a straight rock cover of the source until original vocals done through Ace Studio AI come in at :35, sounding like Nickelback but having some roughness that exposes them as AI, at least on headphones. They sound fairly solid, all things considered. Dynamically, this feels like they have one main style of delivery, so that along with the relatively straightforward drum patterns makes this song feel like it basically has one gear and very samey sections. Hemophiliac had pointed out to me that the drums lack clarity, which also makes them appear generated. 3:09 finally had a dropoff and rebuild, though I'm listening to these lyrics and it all feels very generated before another iteration of the chorus at 3:50. Ugh at "Yoshi" not being pronounced right at 4:28. It sounds like this is built out over the chord progression of the original, but not directly arranging it for extended periods of time, which definitely makes this feel like it's generated music extrapolated from the one segment of the source that's directly referenced. Lyrically, I was initially feeling these could have been made by a human, due to so many different words invoking various Zelda game titles, but having listened to a decent clip of AI-prompt generated music ahead of voting on this, it feels like one extended lyrical word salad about an annoyed Link, and lots of antiseptic rhymes. It does at least have decent prosody to try and make things flow. Sorry if my perception of this undersells how much actual human-generated content is there, Craig. If I had to bet on it though, this comes across like prompt-generated lyrics and the song structure isn't dynamic enough or connected enough to the original VGM to feel like it's mostly Craig's direct input on this. I definitely don't want a trend of people sending AI-generated content here, no matter how good it ends up being. The aim is to highlight skill, intention, and creativity with human-created, human-written, human-produced music. When you take human decision-making out of the equation, even if the end result sounds good, it wouldn't be people genuinely creating the music.
prophetik music ⚖️ Posted Monday at 12:29 PM Posted Monday at 12:29 PM (edited) i'll note that the 'arranger' explicitly avoided responding about who actually wrote the arrangement. they stated who played the instruments but did not clarify who wrote the overall track's arrangement. the track doesn't have enough source so it's a definite no. separately the text is word salad, like larry said (although the joke about going to call mario at the end is admittedly funny), and all of the overheads for the drums are just static. the guitar work is solid throughout, i'll note, and i liked the aggressiveness of the opening section and specifically the altered chord at 0:24 was a fun idea. it's too bad the rest doesn't have anything to do with zelda. NO Edited Monday at 12:30 PM by prophetik music
Hemophiliac ⚖️ Posted 17 hours ago Posted 17 hours ago Clearly there is not enough source usage in this track, it's an original track with a couple quotes from Zelda. Liontamer's breakdown is accurate. By itself that is no. Even if you are being truthful that the vocals are Ace Studio, they fall into the uncanny valley and are not realistic. It took time to add in breath sounds, but there's a few times where a word will blend right into the next word without a break. No time was taken to craft them into a believable performance. The vocals sound over-processed and super pitch corrected into an inhuman perfection. The percussion sounds over compressed and cymbals are sloshy. NO Now, I want to address the question I had previously asked you in the submission thread. At that time you said that the vocals were generated by Ace Studio. This is a non-generative AI virtual instrument. Our policy does allow for that. However, on your YouTube upload you state that they were Suno AI. Suno IS generative AI, and is not allowed. So either you've willfully lied to us and were disingenuous, or you really don't care about what you're doing and don't care about AI theft. Both avenues are not a good look. I personally think you're not being truthful and did not perform anything in this, it sounds entirely generated to me. From vocals, drums, bass, guitars, everything. If I'm wrong, and if that's the case I'll be glad to eat crow. However, they all still sound unreal and inhuman as well as poorly produced. Please rethink your approach to music if you consider submitting to OCR. This is not the place for AI content.
paradiddlesjosh ⚖️ Posted 16 hours ago Posted 16 hours ago (edited) Craig, this is going to sound blunt, even rude. You might feel insulted. The lack of identifiable source material usage alone would be enough to keep this track from passing the panel. Two brief, straightforward quotes from "Overworld" and half a verse's worth of vocal melody across a 4-minute song do not a ReMix make and no amount of goofy meta-narrative lyrics will change that. And frankly, it's questionable how much of this, if any, is your production. If the drums are indeed GarageBand Drummer, they need more humanizing (velocity variations, nuance, ghost notes, etc) so that the part doesn't sound like a string of loops from a MIDI pack. If the vocals are truly Ace Studio and not a hallucination from Suno, then this, too, needs more effort on your part to craft a believable performance, as Hemo mentioned. This is without mentioning the limiter settings or the warbly shmear on the instruments and vocals. That's all a hard NO right there. Assuming you made this track as you claimed to have made it in your comments on the submission, it needs a massive overhaul before it's ready to be posted here. We have no issue with the use of virtual instruments or effects that utilize machine learning (e.g. SWAM, Ace Studio, iZotope Ozone), provided the track's execution conforms to our Submission Standards (which I encourage you to read and familiarize yourself with if you wish to submit to us again in the future). But we don't allow generative AI elements in tracks at OverClocked ReMix, full stop. We're here to celebrate VGM as a human-created art form, not merely to provide a platform for content. HELL TO THE NO Edited 6 hours ago by paradiddlesjosh
Chimpazilla ⚖️ Posted 3 hours ago Posted 3 hours ago Cosigning with my fellow Js. Cool concept, and the composition and mix sound good enough to do very well on YouTube, Spotify and all the other socials, but for OCR we look for very specific things, including enough source use with thoughtful and creative interpretations, and real musicianship. While we allow limited use of AI tools to enhance a track, this submission (and the other tracks you submitted on the same day) is far too much AI for our standards. If we have misjudged this situation, I'd welcome some real, solid proof that these tracks were made with a lot more human involvement than we are hearing, but my feeling is that such proof cannot be provided. Best of luck to you with these, you've hopped on a shiny new lucrative bandwagon with these AI tracks, we just don't want to host music like this on OCR. NO
pixelseph ⚖️ Posted 3 hours ago Posted 3 hours ago As much as I'd like to see the human involvement in this piece or the other submissions by Craig, it is overshadowed by the audacity of submitting an entire album's worth of generative AI slop and lying about how large a role generative AI played in it. I have no pithy comments to make - my fellow judges summed up my feelings pretty well, and there's no need to belabor the point. NO. DO NOT resubmit anything using Suno or other such tools again and lie about it.
Recommended Posts