- Видео 165
- Просмотров 12 626
Twobob Club
Добавлен 10 янв 2008
The archives of Twobob
digital tuna (x.com) - Less than 512MB and 130 seconds mix
Twobob_bits
x.com/Twobob_bits/status/1784379063479472192
x.com/Twobob_bits/status/1784379063479472192
Просмотров: 370
Видео
welcome to the club [gettin messy mix] alt video cut
Просмотров 3164 месяца назад
alt video, with a view to the real
welcome to the club [mambo jumbo mix]
Просмотров 2314 месяца назад
welcome to the club [mambo jumbo mix]
welcome to the club [gettin messy mix]
Просмотров 3934 месяца назад
welcome to the club [gettin messy mix]
Outside the club - free Jazzy classical AI explorations
Просмотров 114 месяца назад
Exploring elsewhere, that is not a beat, or maybe this time obvious thing is obvious, maybe it is not.
Welcome to the club
Просмотров 3944 месяца назад
because @ChasingTheGoose was messing about. Video is custom wangfuyun/AnimateLCM script / audio Udio. Demucs 4 or 6 deconstruction is a nice parts balance sculpting and beat / vox / other rework station workflow Individual samples can be generated using the public free harmonai scripts and of course some APIS some fun
infinite spaces between the infinite spaces in infinitely spatial space
Просмотров 394 месяца назад
infinite spaces between the infinite spaces in infinitely spatial space
HarmonAI Dance Diffusion Fine-Tuning Howto
Просмотров 3,4 тыс.Год назад
A walkthrough of the Dance Diffusion fine-tuning script. NOTEBOOK: colab.research.google.com/github/Harmonai-org/sample-generator/blob/main/Finetune_Dance_Diffusion.ipynb GITHUB: github.com/Harmonai-org/sample-generator WANDB API KEY: wandb.ai/settings CHUNK HOWTO: ruclips.net/video/WtSBpkXK-0A/видео.html WGET FOR MODEL: gist.github.com/twobob/b9a46be801598fb89cc2e058cfbe38e5
HarmonAI Dance Diffusion AEIOU chunking script Howto
Просмотров 1,1 тыс.Год назад
DISCORD: discord.gg/harmonaiplay COLAB: colab.research.google.com/ Typical settings for chunking .wavs ready to use in fine tuning a 65536 sample size Dance Diffusion model. AEIOU: github.com/drscotthawley/aeiou For more detail on the individual settings please see the discussion on the Discord here discord.com/channels/1001555636569509948/1001555639228702822/1054859075986395177
HarmonAI Dance Diffusion Inference Howto
Просмотров 3,1 тыс.Год назад
Quick step by step video for doing inference using the Harmonai system Under 5 mins be ready to make some sounds... GITHUB repo of Harmonai-org github.com/Harmonai-org/sample-generator
A Place Of Our Own Feat Textured Grooves
Просмотров 13 года назад
A Place Of Our Own Feat Textured Grooves
Golden Shard Of Light Pablo Almazan drosophila ultravioleta
Просмотров 33 года назад
Golden Shard Of Light Pablo Almazan drosophila ultravioleta
the high end percussion is noticeably hissy or gassy, I wonder if its possible to get cleaner sounds. is the track made from individual stem prompts or just one big prompt?
just one big prompt on this one. it was the inception of single prompt to audio. that said: could rip it to pieces with stemming tools and rework the hats in a pinch. :)
@@twobob thanks. I need to figure out how to use the software, only came across it today
@@spoony01428 okay. here's another thing then using musicgen, and a seed audio track gist.github.com/twobob/f4b9fb0b42e9a4f1128ded662f0fb41b That was kinda fun to use. Managed to get it to 860 seconds on a 3090. Not quick. And again would need running over with demucs. See line 39 for a howto demucs.separate.main(shlex.split(f'-n mdx_extra --shifts 1 --overlap 0.15 --float32 --clip-mode clamp \'{file_path}\'.wav'))
Oooo I’m feeling it 🎉🎉
2:30 - 3:00 begin dental descent
Thank you! I've been trying to access this model for a long time since the demo went down, and I've finally been able to use it again, thanks a lot.
cool
Hi there , Im a complete newbie to all this . Im trying to follow the how to.. but I keep getting a blank screen with a Loading box in the bottom left when I go into google colab ? ive registered a drive account and weights and biases . Not sure whats going on really. Im on a Mac OS 10.16.3 , so its pretty old . Is that the problem do you think ? Any help appreciated thanks.
Im guessing your browser is the issue?
@@twobob I’m gonna try it later, thanks for replying.
@@twobob it works! Thanks. I’m feeling real dumb now ☺️ . Will probably be pestering you in the near future. Exciting isn’t it ? Not the pestering part , ai I mean . 😃
is there no way to run this without google colab? like, fully offline
just download the files. adjust the paths to work locally. install the deps in the python environment you are going to use and rip out the colab bits of the main loop. Totally doable. Probably could ask chat gpt to do that
When I tried to run the last part, it gave me a green tick instead of continuing to run, and my project on W&B says "Failed". I have already done the chunking process, so is there anything else I could be doing wrong? Thanks!
link to your files maybe. That sounds like the error in Logs on W&B (might glean something) - I recall there is a text version somewhere in the runs. WANDB is down atm so I cant test . sorry
Thanks so much for the quick response, I didn't have one of the models on my drive XD but it's up and running now. Seriously, a gigantic size thanks for all your work on these tutorials, I'm having a blast playing with this thing @@twobob
colab.research.google.com/drive/1C4fPoRMimtVJ23gi9TCIrQTYW29_NNmE?usp=sharing Seems it needed a massage
Hi, thanks very much for making these videos! When I run this colab, I get this error: ''ar' object has no attribute 'nopad'. No chunks are created. I searched the discord and found other people having this same problem, but I didn't find any solutions. Is there anything you might suggest?
github.com/drscotthawley/aeiou/pull/34
colab.research.google.com/drive/1C4fPoRMimtVJ23gi9TCIrQTYW29_NNmE?usp=sharing enjoy
dude i'd pay you to read a book for an audio book your voice is perfect. also thanks for explaining it. got a bit of trouble on how to use this
I love your vids but man am i sleepy now. voice is so soothing lol
hah. next time may I go with "advertising shouty man"
mans is whispering
will knock up the levels in future vids
All that work and you didn't even have the courtesy to play the tracks at the end! Urgh, it's so frustrating. Lol, not really. Thanks for the info
eh. didn't miss nothing
I have volume cranked up to 100% and still have tremendous difficulty trying to hear what you're saying. Such a shame because this is nonetheless an amazing video.
okay. i can provide a loud version if you need it. or the automatic subtitles should work
@@twobob appreciate it. i actually did get a lot from the video, my comment was more like a suggestion on what you could improve. love and peace to you.
@@shfunky and taken as such. AFAIK youtube doesnt allow reposts. The volume was auto levelled in BlackMagic Resolve. next time the settings will be higher, thanks for the feedback, I do recognise that amidst this sea of compressed offerings something less frantic loses the noise war ;)
Liked and subscribed!
Bob Ross of AI
Perm., not included
These videos are extremely helpful Twobob, thank you so much for uploading 🙏🏻🙏🏻 I’m a musician who’s incredibly excited about the possibilities of diffusion in the audio space but I have no prior experience with coding, using github, google colab, etc. So watching you go through these things has helped me a lot! I’ll see you in the discord 🤘🏻
Happy to help. Will keep an eye out
Fantastic
Thanks man. Lots of love into this
you click the "+CODE" button to add code you can run yourself
you click the "+CODE" button to add code you can run yourself
the demos of my model sound great but while generating audios on this notebook it sounds weird, like slowed down and more deeper, I'm guessing it's because i need more training? but I'm not sure because as I said the demos sound great EDIT: I think I found the issue, it was something dumb :P
Hello, thanks for taking the time of making these videos as they are really helpfull for me, a completely begginer, but as the beginner that I am, I have some doubts, my main doubt is, is it okay to have each song separately on wav files like a full album where each song of it is a separated wav, and if so, do I have to ensure myself that there aren't any moments of silence? I think that could affect my final model, correct me if I'm wrong
Hi. This script removes optionally removes silent chunks. turn STRIP to True for this to happen.
@@twobob Nice, and thanks for the response :) I'll be trying it out soon
github.com/twobob/Hobo/blob/main/mp3/8xIPLMS.wav result wav to download
Brilliant video to your great tune.
Great song bro. Subscribed.
That phat bass! Heartfelt Christmas Glee to you and yours.
Love it!
Thanks!!
Well done my friend!
TY gooni
nice map skin. it is from battle pass reward?
was from being a beta user
Is this a game or a dota gamemode? Never heard of this.
game :)
Still playing? Excellent! I've been jumping on now and then recently too, hope you are well TwoBob :)
Same to you fella
Didn't even edit out the crash. Hard core. the real experience.
Really trippy, captivated me & was hypnotic.
😁
Twobob! What a pleasure, hope you well, cool vid
Cheers Si. Same mate
This sounds great!
Cheers Ryan. Some studio time in this one :)
😁
😁
Really beautiful & uplifting
Very pleasant sounds!
Nice one, feels expansive
Suspenseful asf) I love this well done
really nice! great work
;) thanks
great vid man
nice track! :)
This is great!
Awesome tune! keep it up!!!
voted and just wanna say amazing and the only one with imagination and ˝balls˝... congrats ;)