>What is this?A local open weights music generator, like Suno and Udio.>Original repo (includes lora training)https://github.com/ace-step/ACE-Step-1.5>Comfyui guide (but use the SFT model instead of Turbo, CFG=1 and 50 steps)https://docs.comfy.org/tutorials/audio/ace-step/ace-step-v1-5>Suno-like UIshttps://github.com/fspecii/ace-step-uihttps://github.com/roblaughter/ace-step-studio>WIPhttps://github.com/FizzleDorf/AniStudio>Cover and Edit modeshttps://github.com/ryanontheinside/ComfyUI_RyanOnTheInside/tree/main/examples/ace1.5>Cover and reference song tutorialhttps://www.youtube.com/watch?v=sv4pNrjRh7s [Open]Share your gens and lora results.Keywords: music gen, local model, song gen, suno, udio, acestep, ace step, lmg,ldg, dmpPrev: >>108164777
Sharing Britney Spears lora again:https://voca.ro/16blE7la2Ff8https://voca.ro/1eLHswbE9ZHKhttps://voca.ro/1me4VBIkzHfK
>>108195231when are you making an acestep.cpp front end?
>>108195243Not Ani, btw but he is working on that.>>108194165>>108194014
>>108195260Thanks! Suck a cool dev.
>>108195260https://github.com/ServeurpersoCom/acestep.cpp
>>108195295Nonononono Ani please make a wraper!
>>108195231>>WIP>https://github.com/FizzleDorf/AniStudioNice
>>108195231you're literal human garbage julien
So this is the power of remixhttps://voca.ro/192SgYw5zdnz
>>108195231Any music generated with this tool is Digital Fusion. Look up "digital fusion" online to see what i mean.
https://youtu.be/UH47nkLtOcswhen do local get something like this?
>>108196310why do the mods allow it?
>>108198057catjak is a mod and can say anything she wants. she sometimes posts with a vpn pretending to be ani then deletes the posts
>>108195238but where is the link senpai?
>>108198185>she
why didnt it work
>>108199138https://github.com/ace-step/ACE-Step-1.5/blob/main/acestep/genres_vocab.txtandhttps://github.com/ace-step/ACE-Step-1.5/blob/main/docs/en/Tutorial.md
I thought about the legal issues around sharing LoRA the other day, and I don’t think it’s really a problem that open weights have to deal with. What we need is a tool that makes training LoRA easy; something like a smarter and smaller model that can handle the hyperparameter tuning for us. The ultimate goal is to lower the barrier to entry for gen music.
>>108199706just name it generically.
https://files.catbox.moe/kzc8k8.mp3
>>108199706So, audio is very complex, but as both a music producer and a ml enthusiast, there are ways to do this. Do people really want this? Should i make Side-Step do this? I thought the best was granularity and control because no dataset is ever equal.
