Dreaming Tulpa
Dreaming Tulpa
  • 38
  • 98 358
About Coffee - A Lost Twin Peaks Tape of Agent Dale Cooper
In this video, Dale Cooper takes us on a journey to explore the world of coffee, its aroma, and the way it can transport us to different places and times. As we listen to Cooper's soothing voice, we can almost smell the rich aroma of freshly brewed coffee in the air.
This is a AI experiment combining cutting edge (as of time of publishing, this will highly likely already be outdated in a few days/weeks) AI tech and a testament to the power of machine learning and natural language processing, allowing us to generate new dialogue that captures the essence of beloved characters like Dale Cooper's unique personality and style.
For the script I used ChatGPT. For the voice cloning I used ElevenLabs. And for the video footage I used the first publicly available text-to-video model called ModelScope.
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
SUPPORT THE CHANNEL
📰 Subscribe to my weekly AI Art newsletter: aiartweekly.com
🐤 Follow me on Twitter: dreamingtulpa
☕️ Buy me a coffee: ko-fi.com/dreamingtulpa
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
#texttovideo #texttospeech #aivideo #twinpeaks
Переглядів: 788

Відео

The Winter Sounds - Meteor [Alternate AI Music Video]
Переглядів 984Рік тому
My first Deforum Stable Diffusion v2 experiment. I rendered this video quite a few times (around 10 times, each run took around 6-20 hours to render) and combined different segments from five different versions into a final cut this time around. This was also my first music video created upon request by the band @thewintersounds (go give them some love ❤️). The song 'Meteor' is off our 2018 rel...
How to Setup Automatic1111 WebUI with Deforum Stable Diffusion on Paperspace
Переглядів 10 тис.Рік тому
In this video we'll setup the Automatic1111 webui with the Deforum Stable Diffusion extension on paperspace.com. Deforum is a notebook that allows us to create animation videos with Stable Diffusion and it's the one I've been using to create all my music videos: ua-cam.com/video/w9109ULX6zs/v-deo.html. After spending almost 500 computing units on my latest music video experiment (Puscifer - Mom...
Puscifer - Momma Sed [AI Music Video]
Переглядів 4,6 тис.Рік тому
My latest Deforum Stable Diffusion experiment. Basically blew an entire month worth of computing units into this and reiterated around 10 times until I was pleased with the results. I still find stuff that I could've done better, but enough is enough. Prompts are indeed the most important part of the equation here. If you're a fan of Puscifer and acoustic rock music, then you'll love this music...
How to Generate AI Music with Mubert Text to Audio Tutorial
Переглядів 48 тис.Рік тому
In this video we're taking a look at Mubert's text2audio API that lets us generate audio tracks using text. Those track are all completely new and you can use them in your streams and videos without getting hit with a copyright strike or DMCA takedown. UPDATE: After I recorded the video, Mubert updated their GitHub readme. Sounds are not created directly based on the text prompt, they're first ...
The Black Keys - Lonely Boy [AI Music Video]
Переглядів 1,4 тис.Рік тому
I'm still tuning my beat recognition algorithm. This time combining RL footage with AI generated visuals. Original Video: ua-cam.com/video/a_426RiwST8/v-deo.html ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ SOCIAL MEDIA LINKS! 🐤 Follow me on Twitter: dreamingtulpa 🌄 Follow me on Instagram: dreamingtulpa ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ #stablediffusion #musicvideo #experiment
VIDEO ROTOSCOPING WITH STABLE DIFFUSION?! How to create AI Video Input Masks for Deforum
Переглядів 18 тис.Рік тому
Stable diffusion? Nah, that's the scramble suit from the 2006 adaptation of "A Scanner Darkly". Or is it? In this tutorial we're taking a look at how to generate video input masks to only diffuse parts of a video frame that we want to create a rotoscope effect. For this to work you're going to need two things: 1. A runwayml.com account. 2. A Google account to access Google Drive and Google Cola...
Prithvi & drkmnd - Gloom (AI Lofi Music Video)
Переглядів 6 тис.Рік тому
A new Deforum Stable Diffusion AI music video animation experiment. I made further improvements to my beat detection algorithm to make the visual audio reactive. This time adding movement to the scene by reacting to the piano keys and switching the scene based on the beat strength. This time I didn't go crazy on the prompts and just added 41 prompts to match the movement from the piano keys. Ho...
Bonobo - Kerala (AI animation with 238 prompts)
Переглядів 3,6 тис.Рік тому
A new Deforum Stable Diffusion AI music video animation experiment. I made further improvements to my beat detection algorithm to make the visual audio reactive. This time switching the scene based on the beat strength, which for my taste produces super cool results. As it's my first video that doesn't rely on lyrics, I used 238 prompts from Midjourney's community feed to create the different v...
Cyberpunk: Edgerunners - I Really Want to Stay at Your House [AI Music Video]
Переглядів 911Рік тому
Cyberpunk: Edgerunners - I Really Want to Stay at Your House [AI Music Video]
River Tiber - The Star Falls [AI Video + Audio Reactive]
Переглядів 239Рік тому
River Tiber - The Star Falls [AI Video Audio Reactive]
I Love You More - but every frame is an AI image (audio reactive)
Переглядів 295Рік тому
I Love You More - but every frame is an AI image (audio reactive)
La Vita - but every frame is an AI generated image based on the songs lyrics
Переглядів 457Рік тому
La Vita - but every frame is an AI generated image based on the songs lyrics
Joji - Glimpse of Us - but every frame is an AI generated image
Переглядів 384Рік тому
Joji - Glimpse of Us - but every frame is an AI generated image

КОМЕНТАРІ

  • @user-yi4lp2cq1q
    @user-yi4lp2cq1q 3 місяці тому

    How to Setup fooocus WebUI with Deforum Stable Diffusion on Paperspace please

  • @ChillHope24
    @ChillHope24 3 місяці тому

    Where is now the loop mode? :( I'm a Pro plan user, didn't see that yet

  • @lewing-alt1551
    @lewing-alt1551 4 місяці тому

    criminally underrated

  • @TheoreticallyMedia
    @TheoreticallyMedia 6 місяців тому

    I forgot that you did this, and it also reminded me why you're one of my favorite people in our weird little world. That gum you like is going to come back in style.

  • @hajotheunissen5342
    @hajotheunissen5342 6 місяців тому

    Truly amazing! Speechless. A true master in the AI arts...

  • @stereosolsticemusic
    @stereosolsticemusic 6 місяців тому

    very tastefully done visuals and a great song too - a really good combination 👍

  • @ItJustDre
    @ItJustDre 6 місяців тому

    Song gives me Nostalgia of END OF WATCH

  • @agidotexe7167
    @agidotexe7167 6 місяців тому

    Probably the best ai video ive seen, i mean its not saying much cuz its just an algorithm but still neat!

  • @OwaOwaBiowa
    @OwaOwaBiowa 9 місяців тому

    result it's not good , try decreasing denoising strength

  • @lucastorrez4277
    @lucastorrez4277 10 місяців тому

    the deforum tab simply doesn't show up in the UI, even though i see it was installed correctly in the extensions folder :/

  • @drcoprofeich
    @drcoprofeich 10 місяців тому

    great video I was encouraged to try it but I found nothing but problems, after days of trying I think I'm going to give up. The first problems it gave me were with python, using a container I was able to solve this but then others appeared and I could never generate anything in deforum, only stable diffusion works for me

  • @tomsebanc6951
    @tomsebanc6951 11 місяців тому

    It keeps saying I need python 10 or newer. How do you change which version of python it's running? (I have 11 on my mac)

  • @pwnomega4562
    @pwnomega4562 11 місяців тому

    Ya'll cant just sit down and learn how to make music? Its really not hard.

  • @julss6635
    @julss6635 11 місяців тому

    how do you subtitle?

  • @RoboMagician
    @RoboMagician Рік тому

    is there a stable diffusion extension for rotoscoping?

  • @zes7215
    @zes7215 Рік тому

    no such thing as daux or etc, ceouxuax, outx, cna outx any nmw s perfx

  • @creatorsmafia
    @creatorsmafia Рік тому

    ! I never thought AI could generate such incredible music. The variety of genres and the ability to create loopable tracks is impressive.

  • @cl5wless
    @cl5wless Рік тому

    I tried it and it doesnt work anymore idk.

  • @backroomss
    @backroomss Рік тому

    is there any way to get an api? free one

  • @beebop90
    @beebop90 Рік тому

    Wow. Wow

  • @arkelss4
    @arkelss4 Рік тому

    Free colab , was a good run

  • @iREM0015
    @iREM0015 Рік тому

    Is it possible to install lora and the models and vae in Paperspace?

  • @anotherstarwarsvictim
    @anotherstarwarsvictim Рік тому

    do you pay for that?

  • @AlexC-O_O
    @AlexC-O_O Рік тому

    Why use gradient and not CORE? What s the difference??

  • @milestrombley1466
    @milestrombley1466 Рік тому

    I keep getting an error when I try to install the requirements. I am using stable diffusion Voldemort with CPU. For some reason it's not getting pytorch. Is it because I am using CPU in paperspace?

  • @dillonjones2312
    @dillonjones2312 Рік тому

    Which A.I. is this?

  • @yuruhuwa-music
    @yuruhuwa-music Рік тому

    I am a Japanese AI music creator. Thanks for the very interesting video.

  • @user-lv4ov5rp6d
    @user-lv4ov5rp6d Рік тому

    Is Mubert copyright free to use on YT and to sell?

  • @marcelocosta357
    @marcelocosta357 Рік тому

    Como gerar as musicas??

  • @mustafadheyab6513
    @mustafadheyab6513 Рік тому

    Can we generate sound effect with this?

  • @edb8139
    @edb8139 Рік тому

    This video would be more interesting if some AI effects were applied to the person dancing too, and perhaps had them interacting more with the background (using AI).

  • @americanvista
    @americanvista Рік тому

    Sample of AI music, AI creates epic UA-cam soundtrack - no human skills needed! ua-cam.com/video/WzQU_j-ZE8w/v-deo.html

  • @fenix20075
    @fenix20075 Рік тому

    No it is not good... I type orchestra it give me piano, I type in 8-bit and it give me techno. It just like randoming a music to the client and it didn't match the prompts even. The final try I accepted are "psycho techno sad dark hate battle". Those prompts can be referenced from the generated music.

  • @user-pc7ef5sb6x
    @user-pc7ef5sb6x Рік тому

    More baby, more!!!

  • @dreamingtulpa
    @dreamingtulpa Рік тому

    Unfortunately the ModelScope model is overtrained on Shutterstock footage, which is why the watermark is clearly visible in most scenes.

    • @nocturne6320
      @nocturne6320 Рік тому

      Is there a way of generating longer clips, or are you stuck with only these very short ones?

    • @dreamingtulpa
      @dreamingtulpa Рік тому

      @@nocturne6320depends a bit on the GPU. I managed to create up to 48 frames until I ran into out of memory issues. I think the max is somewhere around 128 frames right now.

    • @Dale_Blackburn
      @Dale_Blackburn 4 місяці тому

      Hello bro, can you please tell me how did you do this? I want to add a dreamy sequence to my mid length festival feature film. Can i make an AI mix of my film with a program like you used? Can you please please tell me how can i do this? I don't know anything about AI. I will add your name to my film if you want in the credits, i want to make it look big.

  • @MilesBellas
    @MilesBellas Рік тому

    via koiboi

  • @user-hb6dd9iu9g
    @user-hb6dd9iu9g Рік тому

    💥💥💥💥

  • @GuillaumeGamer
    @GuillaumeGamer Рік тому

    This is awesome, i have a lot of fun using the colab. Could you please make a video showing how to use MusicLM?

    • @backroomss
      @backroomss Рік тому

      how could you use It ? looks like it needs Api and it is not free

  • @user-hb6dd9iu9g
    @user-hb6dd9iu9g Рік тому

    Thank you for this video. Is this colab is free to use?

  • @RobertJene
    @RobertJene Рік тому

    OK how do you do this WITHOUT the apple software

  • @karsoverheaven7970
    @karsoverheaven7970 Рік тому

    I was wondering why it is named like the indian state im from but the animation and the music is soo clean

  • @Olav3D
    @Olav3D Рік тому

    It is paid now…

  • @eiriksundmark
    @eiriksundmark Рік тому

    That was SICK dude! 🔥

  • @MrInterpriser
    @MrInterpriser Рік тому

    Can it generate sound effects based on the prompt?

  • @pkay3399
    @pkay3399 Рік тому

    Can we use it on Paperspace?

  • @oneanswer514
    @oneanswer514 Рік тому

    hello is this possible local install?

  • @ndubuisicharles559
    @ndubuisicharles559 Рік тому

    How do I bypass the audio water mark

  • @thenew4559
    @thenew4559 Рік тому

    When I tried to generate music with the google collab like in this video, it now says that I have to subscribe on Mubert's website to access it :(

  • @splendev3615
    @splendev3615 Рік тому

    this is a paid plain scam not a tutorial for building your own music AI, no thanks

  • @Nekotico
    @Nekotico Рік тому

    not working anymore, its paid now, sad