[ home / bans / all ] [ qa / jp / sum ] [ maho ] [ f / ec ] [ b / poll ] [ tv / bann ] [ toggle-new / tab ]

/qa/ - Questions and Answers

Questions and Answers about QA

New Reply

Options
Comment
File
Whitelist Token
Spoiler
Password (For file deletion.)
Markup tags exist for bold, itallics, header, spoiler etc. as listed in " [options] > View Formatting "


[Return] [Bottom] [Catalog]

File:eb102131812631b187dc3a6c12….png (940.4 KB,1464x1000)

 No.66481[View All]

i've seen dedicated drawthreads and writethreads here, but never once saw a music creation thread. considering you can post sound files here i wonder why it happens, because i did see several posters here said they made music. is it because many consider the stuffs they make have no merit to be posted? or because many have some musician profiles that they want no connection to this place...?
anyway, post some musics you made!
115 posts and 41 image replies omitted. Click reply to view.

 No.98915

File:untitled2.ogg (1.83 MB)

>>98908
Thanks

>>98906
Added a third section and change some things I didn't like.

 No.101401

File:example1.mp3 (811.63 KB)

finally got around to doing a test run, had an idea that I thought needed to be more than one track so I decided to throw this together to see if it would be easy to do. I ended up liking it despite its imperfections, so I wanted to share it

 No.103745

I watched a youtube video about breakbeats that suggested I precure some not sold anymore 90s samplers from the tubes and I looked and there is this huge collection on archive! Exciting

https://archive.org/download/90s-sample-cds

 No.103750

>>101401
Feel like this could be touched up a bit into something really nice. Can sorta hear things in it that are really pleasing to the ear.

 No.103777

>>103750
¥Feel like this could be touched up a bit into something really nice
thats the plan, I finally dumped the money into recording equipment and once I get it figured out I wanted to make another improved and expanded version of that.
I set it up yesterday and was having an issue where the input and output was only coming from the left, I'm going to try to fix that tonight.

 No.103928

Oh kissu, this is it! my magnum opus, my absolute best (for now)! I am SO proud

 No.103946

>>103928
Personally, I feel 0:45-1:05 is a bit too "harsh" for me and a to a lesser extent so is 2:05-2:40, but 2:50 onwards? Good shit, love that part. I'm musically illiterate so I don't know how to express it, I don't know how you did it but I like it, and I can see the song does a lot of stuff in its different sections.
>>103777
trips

 No.103947

also nice skull love skulls wish i had a skull

 No.103951

>>103928
That's very impressive!

 No.103953

File:1744d4b3db8f5e39f88f8c3dfe….png (635.96 KB,1024x633)

>>103928
the sound textures are very nice, the composition is also good
the mixing overall is quite high-heavy and not very ear pleasing, it sounds like 5 layers of OTT stacked on every single track

 No.103954

>>103928
Nice. The visualization is really cool too.

 No.103955

File:[MoyaiSubs] Mewkledreamy -….jpg (413.57 KB,1920x1080)

>>103946
I agree with this guy a bit with the "harshness". I've always been overly sensitive to this kind of thing and it's not a genre I listen to as a result.
I liked 1:20ish to 2:08 as I honestly can't listen to the rough parts without developing a headache.
Congratulations on creating it!

 No.103968

File:tumblr_28f515294a2a36e1ebd….jpg (39.81 KB,538x550)

>>103955
>>103946
>>103947
>>103951
>>103953
>>103954
Thank you kissu, I really appreciate your comments! I'm sorry if I hurt your eardrums!

 No.104054

File:untitled1.mp3 (200.82 KB)

having alot of issues with my daw and recording software, I switch to audacity which gave me the option to switch to only mono tracks (fixing the one side issue) but audacity isnt very good so it chopped up one of the recordings and was giving me even more trouble with the 3rd track so I cut it. I guess I will give up for now and try again later, eventually i will figure it out and actually be able to do something with all these ideas I come up with

 No.104081

Would be cool to enhance my track with some vocal lines, so I've been playing around with the Synthesizer V for a while. It's certainly a skill to get it to sound great but the interface seems pretty nice. Not that I actually tried any other Vocal Synths hehe. Whats pretty lame is that a purchase apparently gives you three non simultanious licenses, which can get lost if you dont properly deactivate them locally in case of lets say a CPU upgrade. Thats pretty restrictive SIGH hate DRM.

Please share if you have any experience with vocal synthesizers!

 No.104083

File:1531240918319.jpg (8.42 KB,221x228)

Anyone interested in doing a collab song? I mostly do high-energy electronic music. I'm familiar with FL Studio, LMMS, and Famitracker but if you want to work with different software I can try to accommodate.

>>103928
Nice and hard, don't ever tone it down!

 No.104108

File:5b092f21e3dc0264de48674cbd….jpg (74.57 KB,1011x677)

>>104081
The commonly used vocal synthesizers in Japan are Vocaloid, Synthesizer V, CeVIO AI and NEUTRINO.

Vocaloid needs no introduction and you can download cracked version + voice banks from nyaa.
Synthesizer V has a cracked version but voice banks are hard to find.
CeVIO AI has been gaining popularity recently thanks to high quality AI singing, but it's not cracked and voice banks are hard to find.

Thankfully there is a free solution: NEUTRINO is a free AI singing software and 12 voice banks are available for download.
https://studio-neutrino.com
Due to AI singing, the voice needs way less effort to tune and the quality is quite good in the mix, you can search NEUTRINO covers on YouTube to see how it works out.
It has no UI, you need to create MusicXML in Musescore first, and run included batch script on it. If you want a UI, an unofficial UI can be found here:
https://github.com/sigprogramming/tyouseisientool/wiki/Getting-Started

Of course, AI singing technology is constantly evolving, so there are lots of research projects that can generate ever better voices but they aren't very user friendly and often require you to make your own voice banks.

 No.104110

*cough*
>>90430
It'd be nice if someone made use of that thread before it fell off

 No.104137

>>104081
I've used UTAU before, it works. It's pre-AI and just fades pre-recorded phoneme pairs into each other, so it has a bit of a learning curve and requires a lot of fine-tuning. Because it's free there are lots of crappy voicebanks but lots of good ones too. The only thing I hate about it is each voice bank can use a different system, and some of the popular voicebanks are either difficult to use or just plain incomplete and choppy-sounding for English. You know what, don't use UTAU, it's a pain in the ass.

 No.104266

File:R-1677471582988.mp3 (600.61 KB)

Some nonsense I made in BeepBox, mostly to test it out. Seems cool, if a bit limited.

 No.104268

>>104266
Try Jummbus, it's an expanded version of BeepBox - https://jummbus.bitbucket.io/
In the past I would export each track separately and finish mixing in a DAW, with some effort you can get something that sounds decent and a little less flat

 No.104270

>>104268
Looks cool, I'll check it out.

 No.104398

File:test.mp3 (227.75 KB)

It seems all I had to do to fix my problems was use reaper, the one I was using before had some issue where it defaulted to stereo recording which doesnt work properly with windows or something, regardless I have got the recording side of things fully figure out though now its too late for me to be making alot of noise. Glad that watching bocchi earlier inspired me to give recording another try, now I am good to go to put down the track thats been in my head

 No.104449

File:1st try-2.mp3 (2.17 MB)

dekita!!!!!
well at least a poorly mixed rough sketch which is missing bass/keys and the vocal track, but its something!
skip to 1:30 for the actual interesting part since the rest is just the 2 main parts alternating to make up a verse and chorus
I was going to put real people vocals on it from the person who will do the other 2 parts, but I could just do them myself slap miku on it and call it a day.

 No.105689

File:Reaper Screenshot 2023-03-….png (97.36 KB,1611x971)

Just found out that Reaper can output some neat GRAPHS after rendering something!
I wonder if it saves them somewhere? Would be neat to compare graphs for all tracks in an album or ep. Definitely cool to so see how the breakdowns, build ups and climaxes compare in loudness.

>>104449
>from the person who will do the other 2 parts
will you be a band?

 No.106607

Loaded some drum loops into my granular synth, randomized some parameters and let it go! Makes for a super cool sound generator.

 No.106616

File:[MoyaiSubs] Reiwa no Di Gi….jpg (311.36 KB,1920x1080)

>>106607
Those drums kind of sound like rockets being launched.
Out of curiosity, are there other drum types you can mess with? What about Mario Paint sounds? That'd be cool.

 No.106691

>>106616
yeah it loads any samples in common audio formats

 No.116347

I have a 2nd Mini Album now, its called Sucide Machine!
It's inspired by the emotional cycle of elation self overestimation failure frustration and recovery.
it happens to me alot (I hope I'm not alone).
So that's 5 hardcore adjacent tracks in total, and what I have learned is that I don't know how people
make full length albums, this EXHAUSTED me. It wasn't any less painful (I'm exaggerating) than
my first mini album, and somehow it suffers from the same general problems too SIGH.
But I don't wanna think about problems now! onwards!
First track is attached

 No.116348

and I'm making another post just for this track, the last one.
Because it's the best! easily the best part of the album.
It's what makes me confident that the whole exercise was defenitely worth it
in the end. Its kinda more DnB so mabey I just feel this way because I'm not super
aware of what's going on in that space.
But it's definitely a significant track for me, from this album it influenced me the most going forward.

 No.118747

File:2024-01-17 19-05-08.webm (1.2 MB,939x528)

I have at times really wished effect plugins would come with a some basic midi support.
On the simplest level something like having a midi triggered ADSR that controls the wet.
Been wondering why that was such a no go. Without I'm stuck using manual automation,
or doing messy things with sidechain compression. Both I find fairly annoying and finicky.

But there might be a way, using Snapheap.

Snapheap is a sort of modular effects rack that lets you set up custom effects by
supplying effects like compression, delay, reverb etc in their most basic form as submodules.
The same with modulation sources. And guess what it has an ADSR and a midi trigger.

So yay, I can finally see if my idea of midi triggered effects is actually a good idea.
I just gotta make them in snapheap.
As an example, a kick triggered ducker.

Anyway snapheap is really cool and when I got it
it was free but it isn't anymore so that's weird

 No.118775

File:[MoyaiSubs] Mewkledreamy -….jpg (261.44 KB,1920x1080)

>>118747
That's cool.
What's the wet? What's a midi triggered effect?

 No.118840

>>118775
In this case wet is a measure of how much the audio signal is affected by the effect.
MIDI triggered means that the affect gets applied when there are incoming note signals

 No.120563

since suicide machine ended up being very exhausting and only debatably worthwhile
I decided to make a track with a more relaxed and take it easy kinda attitude.
And I'm very pleased with it yes. I should take it easy more often.

In other news, most of my visualizer code now runs on shader. that makes it go in real time, theoretically.
thats cool right?

 No.121923

File:snapSend.gif (786.12 KB,1090x898)

another thing that I like about snapheap is that I can easily set up effects on a parallel track and then have send and return controls for that without making an actual new track in the daw. Reaper can do on track parallel processing natively but I find it is much more convenient to do it in snapheap. Honestly I think all signal length changing effects like delay and reverb should have send and return natively but so far I haven't seen that...

 No.122370

File:Synthetic Escape.mp3 (1.42 MB)

ai music with sono what do you think?
its not overly censored yet I think so uhhh make what you can as long as the model is hot

 No.122384

File:[SubsPlease] Show by Rock!….jpg (110.73 KB,1280x720)

I don't like the genre to begin with so I can't judge it, but it feels extra noisy to me in an almost imperceptible way. I can't remember what it's called, but like the background noise that's really low or something.
AI music seems kinda meh to me so far. The fun will be inserting lyrics, but I don't expect it to sound good. The existing AI covers are alright to laugh at, but I don't think anyone listens to them for sheer enjoyment of the musical arts.
AI's main value to me visually is porn and AI's main value to me with text is porn, so with music... I guess there's joke-y moany doujin music but ehhh.

 No.123272

probably my best investment decision is abandoning art for engineering

 No.123273

haha, crazy. It just straight up ripped off Bach

 No.123274

And the "Dubstep" section is formulaic and vocally a complete copy of Infected Mushroom

 No.123275

>>122384
>>123273
>>123274
Yeah, that's kind of what I was wondering about with this stuff. I've listened to a bit of these and many of them sound 'familiar' but nothing that really stands out to me and wows me as a new and interesting sound. That's probably what makes it good for comedic stuff most.

 No.123278

>>123275
it's good for advertising. All of these AI tools are boiled down to a single concept: Advertise cheaper with similar impact.

 No.123279

creative tools anyways. Even in the case of generating art, people pay for the artist to do it, not for the art.

Pirate minded people don't understand this

 No.123289

File:[Piyoko] Himitsu no AiPri ….jpg (247.91 KB,1920x1080)

Audio AI will likely go through the same process the image and text stuff went through. Absolutely amazing sky-is-the-limit thinking and then people notice all the flaws and patterns and it becomes a lot less interesting over time while the crypto people and investors and CEOs hype it up so they can make money.
Anyway, I don't consider it to be music creation just like prompting images isn't drawing.

 No.123290

>>123289
the on topic sagey

 No.123685

I wonder if composition can actually be done by AI... as with all AI tools I've used so far I find them more of assistants to get you up to speed with the formulas and basic design of art faster than what you'd get without a mentor.

Can it imitate something weird blending 3 art styles into one like this...

 No.123686

Cosmograph's sound is too good

 No.124956

https://www.youtube.com/shorts/ktP-judV7-8?t=15&feature=share good advice. Really is easy in creating audio maximalizsm

 No.135152

File:Games_r1_2024-11-14.flac (10.07 MB)

Created something putting together cuts from https://www.youtube.com/watch?v=WZI1DiyNPUA using Ardour...

 No.135188

File:[Piyoko] Himitsu no AiPri ….png (1.41 MB,1920x1080)

>>135152
Neat! Although, it kind of feels like it's a song that's about to start but never does. This is just my own personal tastes speaking, though.
Well, it might also be my familiarity with the source song, too.




[Return] [Top] [Catalog] [Post a Reply]
Delete Post [ ]

[ home / bans / all ] [ qa / jp / sum ] [ maho ] [ f / ec ] [ b / poll ] [ tv / bann ] [ toggle-new / tab ]