Give a Face to your AI Songs with ComfyUI and FaceFusion
Vložit
- čas přidán 2. 07. 2024
- Let our sponsor BetterHelp connect you to a therapist who can support you - all from the comfort of your own home. Visit betterhelp.com/bobdoylemedia and enjoy a special discount on your first month.
Let's have some fun with animation and music!
Using a song created in Udio, ComfyUI, and Facefusion, we’re going to create a dynamic AI animation.
We’ll use ComfyUI to create our AI performer, driven by a video I shoot with my phone. We apply a claymation effect and do our best to create lip sync!
Udio AI Music Generator: udio.com
Download ComfyUI AND FaceFusion with pinokio.computer
ComfyUI Workflows:
Easy Video Styler Workflow: drive.google.com/file/d/1DGfQ...
Instant ID with Reactor: drive.google.com/file/d/1PPkv...
00:00 Introduction “Performance”
00:35 Channel Growth “Thank you” and New Sponsor Introduction
03:06 Creating AI-Generated Music Videos
03:37 Using Comfy UI for AI Animation
04:47 Extracting and Preparing Audio
05:44 Building the AI Singer
10:33 Animating the AI Singer
18:21 Finalizing and Testing the Animation
25:27 Conclusion and Encouragement to Subscribe
👍 LIKE If you found this video valuable. 🙂
🥰 SHARE If you know someone who might enjoy this video.
⏬ DOWNLOAD or ADD This video to your PLAYLIST for easy access later.
💬 COMMENT Your thoughts and questions are welcome!
📝 SUBSCRIBE / @bobdoylemedia
That's what keeps me going!
🔔 And make sure to hit the NOTIFICATION BELL to stay updated! 🔔
🌎EXPLORE
bobdoylemedia.com
🗓️ MY HISTORY
🧠 meetbobdoyle.com
🤫 Featured Law Of Attraction and Neuroplasticity expert in the book and film, “The Secret”. • The Secret
🎙️ Voice Over Artist 30+ years
📻 Broadcaster / Actor / Creative
❓ASK ME
Got a Question? bobdoyle@me.com
📌 FOLLOW ME
Facebook: / bobdoylemedia
Instagram: / bobdoylemedia
🪙SUPPORT
If you want to support me, the best thing to do is to share the content… sharing is caring!
If you want to support me financially (completely optional and voluntary, but a lot of people have asked for this):
www.thebobdoyleshow.com/tip
Learn more from Bob Doyle Media: www.bobdoylemedia.com - Věda a technologie
Great viseo , love it ! I can see where this leafs down the road and with different models ! Thanks for showing us early!
Love your channel. U can't even imagine how much you've helped us. Ty 4 everything ❤
Thanks so much Bob for this tutorial! Amazing as usual. I'm not sure what the fuss is in the comments about this looking "terrible" this isn't a Disney Pixar animation studio. AI is still in its infancy and these videos are helpful for getting people learning about AI and how to utilise it as the global economy shifts towards a more AI driven economy. These skillset will prevent you from becoming obselete in the near future so rather than complaining about how it looks, start thinking how you can monetise these skillsets. Best of luck.
Ayo ! Mr.Bob dropped Another Gem ! Let's Go !
Thanks so much!
You could also try out the wav2lip node and make the lipsync in ComfyUI 😊
If you swap it twice you’d get better result. You could also select the frame that was not detected and render just those parts. I’d once swap 3 times for a face. You could also include dream machine into the work flow to get a consistent character and add more dynamic to the scene.
Thanks for the tips!
Excellent presentation. Thank you
Bob, could you make a comparison video between ACE Studio and Synthesizer V?
They both have that ‘Acapella Conversion’ feature you spoke about before but ACE Studio has a subscription model and Synthesizer V doesn’t. Just one time price to own the software forever.
Thank you Bob!! 😊
I watch your videos and I do enjoy them.
Really enjoy your creative process. Technology will only improve going forward.
great video
Who knew the SkyNET would have a reference point in the past - a workflow from Bob Doyle...?
maybe you select face BOX and blue, so its kind of getting off the mask
You should work out the problems not just tell followers to tweak. To bad inswapper don’t let us use highres faceswap. It’s often the face enhancer that makes it glitch and go wrong. But either that or lowres face.
Wwwtttttffffv was that xD ....
Why the hell would anyone want to make that?
Nightmare fuel?
lol, ppl in the comments like, why would you wanna make that and it doesn't look good...this has implications well beyon music videos...yall lack imagination...
I'm sorry , I don't wanna be "that guy" but this really looks terrible. It's just better to get a random video and slap it on the music. Absolutely no image coherence, no lipsync, no synchronization between the rhythm and the movements. Literally not a single thing that's worth all the hassle.
Can’t argue with any of that in this case. The claymation effect was a bad example for several reasons, and I didn’t take the audio sync into consideration when I edited out the first part. When it came down to it - and after all the time it took to get there - I just decided to go with it to demonstrate the concept. It probably would have been beneficial to show a few examples where it worked really well. The workflows that would provide that coherence - which I do have - is pretty complex and would take a long time for a beginner to start with.
But I hear ya.
@@BobDoyleMediait's a great demo , and i can see were this leafs diwn the road with a bit more work ! Thanks
Thanks right, but its a great way to learn things and I appreciate a lot, that Bob in the first seconds tells you: "This is what you get!". No hype and cherry picking results like other ai-youtubers.