Destructive Commons.

Don't worry, we'll find you.
Come play with us.
http://www.destructivecommons.net/

Destructive Commons - Britney/Dylan/Java/Survivor/AtomicFireballs meet and play

Victim : Coldplay
Accomplices : Eminem, Le Tigre, Mike Jolly Jackson, Dimu Borgir

Destructive Commons - Coldplay with us

Monkey Business

From the No Human Techno Jazz Orchestra
http://www.no-human.com


Monkey Business [Official no human techno jazz orchestra]

Destructive commons of the "Enter The Ninja" track from Die Antwoord. 
Original ninja voice
New Lion chorus
Instrumental entirely from the Shameless No Human Orchestra

Join us : www.no-human.com && www.destructivecommons.net



D.C. | Die Antwoord - Enter The Ninja (Lion's cover)

Yet another destructive commons :)
Original voice from Jungle Book





D.C. Jungle Book - I wanna be like you (destructive commons)

Special rythmics from Lord General Saelynh (under Doctor SYgr supervision, as usual)
Yet another Destructive Commons (as usual, join us, www.destructivecommons.org(*))
Original voice from Prince Ali
And yup, we loved the original one so we tried being gentle (did we succeed? ask the original song)
http://www.no-human.com as usual


(*) yep we cut the service, back online beginning of february 2016. Shame shame shame

DC | Aladdin, Prince ALI (destructive commons)

Yet another Destructive Commons (as usual, join us, www.destructivecommons.org(*))
Original voice from the girl with the duck face

The extract was not great, so you may hear some splash/wheez sounds with the voice. We chose to do as if we wanted it that way so we coudl still go to parties and have free beers.

Please remember that this is a small game with an xprimental tech. We do not refund bleeding ears, and Quema loves ya.

www.no-human.com as usual

(*) yep we cut the service, back online beginning of february 2016. Shame shame sham

DC | Ice Queen - Let it go (destructive commons)

New Destructive Commons & Neural cover at the same time. Featuring Benjamin Clementine meeting our dear old IQ which, by the way, tries real hard so don't be too harsh with her.

** For Curious & Scientists :
Neural networks were used to generate the final picture, Texture Nets exactly which still are a wonderful tool. Neural networks were used too in order to extract the voice, please look at old videos credits for paper links. Original style is a picture of a mario bross level

** For Normal People
You do not exist, you just want real hard to. This is definitely not the same. We hug you in the brain. Have a drink

** For Not Normal People
You're not so original, but Quema loves you too. Try keeping this reality exactly as you'd have liked to find it. Good ol' reality

** For Mr B and Annoyed people
Actually, we invited Mrs Q as we had a crappy version of the voice. With a good version, we could do something great.

** For Mrs Q and Other Annoyed people
Hey, she looks like she enjoys it, no ? Does loving her imply forcing her to do only what YOU want her to do. Get a vegetable peeler and love him for what he is.

DC | Benjamin Clementine meets da Ice Queen (yep. proud of it we are)

Original voice from Rebirth Brass Band / Kermit Ruffins

Mardi Gras Day [Destructive Commons]

Destructive Commons - Rihanna - Pon De Replay

Voice extracted by neural networks, video clip with neural style, always a huge pleasure to remix such a great lady/

Ok so please bear with me, this is a combination of using Deep Learning (aka Artificial Intelligence) methods for artistic creation. Tools used there are :

* Open Unmix for the source separation making it possible to extract the original voice (https://github.com/sigsep/open-unmix-pytorch). There are three different github projects doing source separation, I have a preference for openUnmix over the Deezer project but did not give the recent Facebook Research project a try.

* Neural Style Transfer (good old) for generating the final video from the original video clip. Implementation here is https://github.com/lengstrom/fast-style-transfer which is great as long as you get used to the parameters (style weight VS content weight)

Enjoy & feel free to make any critics :)

D.C. | Rihanna, Pon de Replay

Destructive Commons - Rihanna - Rude Boy

[Not so sure you need to listen to the lyrics. Waitin for OpenAI or any other one to give us a way to change that too ;/]

Voice extracted by neural networks, video clip with neural style, always a huge pleasure to remix such a great lady/

Ok so please bear with me, this is a combination of using Deep Learning (aka Artificial Intelligence) methods for artistic creation. Tools used there are :

* Open Unmix for the source separation making it possible to extract the original voice (github.com/sigsep/open-unmix-pytorch). There are three different github projects doing source separation, I have a preference for openUnmix over the Deezer project but did not give the recent Facebook Research project a try.

* Neural Style Transfer (good old) for generating the final video from the original video clip. Implementation here is github.com/lengstrom/fast-style-transfer which is great as long as you get used to the parameters (style weight VS content weight)

Enjoy & feel free to make any critics :)

D.C. Rihanna | Rude Boy

Rihanna - Man Down - Destructive Commons

Don't be a sheep, accept the truth. It's been hidden for so long, but now, deep stuff can no longer hide it : Meow miaw! Miaw, merraw meow. Merrawww? Rrrrrr meow meow! Ok, we know, it's a tough one to swallow. We've been lied for so long! Think about it, Merrawwww! And not Meow, as they wanted us to think. Thank you for your attention, please tune your 6G mental chip on our frequency for more.

(It could help : our band is led by a cat. I reckon it can be a bit frustrating if you do not talk cat fluently. Which you shouldn't, considering that you probably belong this low "human" cell group. In cats opinon, humans should not try to understand superior language. Imagine a human purring, it would lead to the end of our world (probably))

For real :  Yet Another Rihanna Cover. Penultimate, probably, but we loved playing with the lady's voice, which is pure magic, swear to a cat. One more and we'll be able to remix the Pokemon theme (....Yeah...)

Voice extracted via cute LSTMs, Open Unmix Nerim style, https://sigsep.github.io/open-unmix/
Instrumental recreated by Quema & Panda
Gabou paid for the stuff, but with interest
Video clip made playing with U^2 Networks (https://github.com/NathanUA/U-2-Net) && Good ol Neural Style (Gatys et al forever)
Pit probably despise the whole graphical stuff as it lacks stones & mud
Lion & Mocking Bird are hunting

D. C.  Rihanna - Man Down

Join the fight agains humanity, bamboo thieves, not-salmon music, frigid lady teddy bears, ignorants, meat and banana communist haters ...
http://www.no-human.com

NoHumanOrchestra

Popular tracks