How to Make Talking AI Avatar (Lip Sync AI): From Beginner to Expert
HTML-код
- Опубликовано: 7 сен 2024
- Learn How to Make Talking AI Avatar with Lip Sync AI, From Beginner to Expert. This guide is perfect for anyone who wants to create talking avatars, use Lip Sync AI, or wants to use SadTalker for making talking AI avatars. We cover the basics of creating AI avatars, including how to create a talking avatar, how to use lip sync AI.
📚 Join our Free and Exclusive Newsletter:
mwealth.link/N...
🚀 Find and Make Viral Videos faster (⭐Highly recommended⭐):
mwealth.link/1...
🛠️ Tools That I use and Recommend:
🎵 Best Music for Videos: mwealth.link/E...
💡 AI Content Creation Tool: mwealth.link/T...
🎙️ AI Voice Generator: mwealth.link/E...
🗣️ Talking AI Avatar: mwealth.link/H...
🔗 Links:
Leonardo AI - AI Image Generator: mwealth.link/L...
Fooocus - AI Image Generator: mwealth.link/F...
ElevenLabs - AI Voice Generator: mwealth.link/E...
SadTalker - Talking AI Avatar: mwealth.link/S...
❤️ ☕️ Support us on Ko-fi! ☕️ ❤️
mwealth.link/T...
🌟 Generate Avatar with Leonardo AI
🖌️ Modify Avatar with Fooocus
📜 Script with ChatGPT or Perplexity AI
🗣️ Voiceover with ElevenLabs
🎬 Animation with SadTalker
#talkingavatar #talkingaiavatar #lipsync
![It took 2 hours to make $220,000 with this AI video [TikTok Shop Affiliate]](/img/1.gif)
Thank you for watching! if you have any question please let me know in the comments👍 🔗 Links:
🚀 Find and Make Viral Videos faster 🚀 (Highly recommended): mwealth.link/1of10
🎵 Best Music for Videos: mwealth.link/EpidemicSound
💡 AI Content Creation: mwealth.link/TuganAI
🎙 AI Voice Generator: mwealth.link/ElevenLabs
🗣 Talking AI Avatar: mwealth.link/HeyGen
Fooocus - AI Image Generator: mwealth.link/Fooocus
SadTalker - Talking AI Avatar: mwealth.link/SadTalker
📚 Join our Free and Exclusive Newsletter: mwealth.link/Newsletter
Dude, you're the only channel on the entire youtube to teach this end to end... Hats Off!
Thank you so much, I really appreciate it
Just discovered your channel, and this great video. best a.i avatar I've seen yet good job well done .
Thank you so much, I appreciate it a lot 👍
OK, these tutorials are just plain fun! And quite easy to understand.
Thank you so much for your kind words! I'm glad you find the tutorials fun and easy to understand. That's exactly what I aim for in my content. 👍
Thank you for these great resources, and in-depth explanations. I hope your channel skyrockets to the moon. Maybe, if you have time, you can create a very short video explaining the hardware requirements for getting these results. I'm sure many are interested in what laptop/computer they need to have in order to run Stable Diffusion at a pro level.
Thank you so much for your kind words! I'm glad you found the resources and explanations helpful. To run Stable Diffusion, you dont need a really powerfull pc, a cheap pc can make it run easly its not a heavy program, the only thing is that it will take more or less time depending on the pc specs. 👍
Your tips are gold! They need loads of practice...but I ll get there!😊
Thank you, please share your journey 👑
if you are going to make conntent i think its way easier to have more monitors
Good luck to all
lets BoooM on 2024/2025 that noone watch news on TV anymore 💪💪💪
Whoa! Never seen the subscribe button in the description box light up!!
Thank you very much for great video
Thank you so much for your support 👍
Hi LEGEND
New subscriber here trying to get along on 2024 with a news channel
maybe a bit late but never to late haha
I love your tutorial
Hopefully gonna make more conntent soon not only shorts news lool
Great, I hope you make great content 👍
Nice video
Congrats for 1k subscribers
BTW I didn't find prompt for image😅
Thank you for watching and for your kind words! I couldn't have reached 1k subscribers without all of you. I'm truly grateful for your support!
Sorry forgot to put the prompt, this was the prompt:
Highly detailed illustration of a photo realistic beautiful girl, golden ratio face, divine proportion, full-body shot, playful body manipulations, rule of thirds:3, pajamas, Masterful brushwork, Vibrant palette, Emotional depth, Dynamic composition, Rich textures, Timeless subject, Subtle lighting, Dramatic contrast, Artistic flair, Refined elegance, vibrant background, full motion effects, diagonal view, crystal particles glittering, non-Duchenne smile, back light, by Skyrn99, ultra sharp focus, vibrant color, Bioluminescence, high quality
Thank you, Mind Wealth, for all these tools, items, and toturial. It's interesting to learned more. Sure I will try and practice it.
Thank you for your appreciation! It means a lot to me that you found value in the tutorials. I hope they help you on your journey! 👍
Wow thank you
No problem, thank you for your support
Your Ai Avatar voice is really nice. Its good to hear.And also thanks for this toturial i have been looking for it so Thank you 😊
I really appreciate your feedback! It's great to know that you enjoyed the AI Avatar voice and found the tutorial useful. Thank YOU for watching and supporting my channel!
@@MintedWealthOfficial You are welcome 😊
I agree. The voice is great. What ai too did you use for your lip-sync in these videos?
Thankyou for the video, It's very quick to understand, and I didn't feel the time of the video, it's really very good.
Thank you so much for your feedback! It means a lot to me that you found the video helpful and engaging. 👍
I just hope you will never remove this video so I can go back to it and try to apply the steps
Don't worry, I won't remove it
I love how AI seems to be somewhat conflicted in 1:50
You told it to make a hacker, and to put her at the desk, which suggest formal attire and emphasizing rather intellectual quality, but you also put in a "red dress", so it tries to make something with a collar, but also with cleavage.
lol yes, thats funny
Very good explanation. In the explanation you explain how to make better sync changing the parameters ein the HuggingFace and Stable Diffusion interface. How can I change the parameters in the Colab version?
Thank you so much! In the Animation section there are this params: --still --preprocess full --enhancer gfpgan, change those to the ones you want, here is the documentation github.com/OpenTalker/SadTalker/blob/main/docs/best_practice.md
Thank you for this interesting video.❤
I'm glad you found the video interesting! Thank you for watching and taking the time to comment.
Cheers for sharing! 🏆
Please consider increase the font size on webpages in your videos. As a viewer I would be quite grateful!
Sure, I will increase the font size
Nice... another great video.
Thank you so much for your kind words! I'm glad you enjoyed the video. 👍
This is a great video!!!
Thank you for the kind words! I really appreciate your support.👍
Since The Free ElevenLabs plan doesn't give commercial use license
will that be a problem later ... a copyright claim or something ???
Yes, you should use the paid version for commercial use
3:37 do you know if it also talks in those languages or is it only for inpu purposes?
Really like your videos btw👍
I also talk in those languages
@@MintedWealthOfficial i mean the output. or is it only english?
i dont understand the last part, how did you make that stable diffusion, can you reaply from start
Stable Difufusion web UI is also called Automatic1111, just install it then follow the instrucitons in this tutorial, its really easy to install just search for it here in RUclips 👍
How to use stable defusion webUI
I dont have a tutorial for that, you must watch another tutorial.
Just discovered your channel, and this great video. Urgent question. Can the same be done using a portrait photo of someone, or must an avatar be used? Looking forward to the answer.
Absolutely! You can definitely use a portrait photo of someone else instead of an avatar. 👍
That's what I was hoping for. Thanks so much for this great video and the super fast reply!@@MintedWealthOfficial
Thank you so much for your kind words! I'm glad the video was useful to you. 👍
my friend how do I run the stable diffusion UI ?? Do I need it to be preinstalled?
Hello my friend, yes you need to install Stable Diffusion UI before, I will make a tutorial about how to install it soon. 👍
Hi. Can you help us to to lip sync our avatars with our own voice file and language other than English? Thanks.
Sure, you can use your own voice, its the same steps, and it works on any language, so dont worry, it will work fine. 👍
@@MintedWealthOfficial Thanks.
when I try to generate with stable diffusion it always show error. Any tips?
What kind of error are you getting?
Which ai voiceover are you using ?
Alex expressive narrator 👍
I tried Clicking on your links for the platforms you used but sadly they did not work.
Could you please tell me which one of them did not work, I just checked all of them and they are working fine.
i got problem running at sd - RuntimeError: Cannot load audio from file: `ffprobe` not found. Please install `ffmpeg` in your system to use non-WAV audio file formats and make sure `ffprobe` is in your PATH. i already installed ffmpeg. any solution?
yes first you have to install ffmpeg in your system, after that make sure to add ffprobe to your path, its really simple just google that "how to add ffprobe to my PATH" it does not take more than 2 minutes. 👍
@@MintedWealthOfficial already installed but still same problem.. but i solved it already there some missing folder in sadtalker. Still cant run it..try test with 3 min speech my pc out of memory it say..when rendering processing just 4% and stuck.. my pc 16ram
Thats weird, I would reinstall everything again, just to make sure that nothing is corrupted 👍
By the last option do you mean the stable diffusion web UI? Do I need to install this on my device to access the screen in your video at 6:46 ?
Yes I mean stable DIffusion UI, yes you need to install it, its really simple and it works faster. To learn to install it just type Install Stable Diffusion UI or Automatic1111 👍
@@MintedWealthOfficial Is this possible on a Macbook?
Yes, follow this instructions: github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Installation-on-Apple-Silicon
@@MintedWealthOfficialis the last option possible on android
How to install sadtalker....???can anyone help me????
This video will show you how to do it: ruclips.net/video/rqIcmzaXepM/видео.html
how open Fooocus ?
You need to install it
What voice are you using?
Alex expressive narrator
Why does my foocus lack 16:9 ratio and gives me error message when generating please?
you have to add that ratio, in the fooocus folder there is a file named "resolutions.json", just add your ratio there. 👍
@@MintedWealthOfficial thank u very much but I can't find that file. Any idea about how to solve the error message when generating? 🙏
which error message are you getting?
In the second way it didnt work to run the seventh setp "visualize code form makeittalk" is that normal?
It could be bussy or something, it never happened to me, if you need more help let me know. 👍
@@MintedWealthOfficial could u share link to do it like u do in thirdway? and... how could i run the same model if i log off and i want to work again with the same image? Thanks a lot in advance! and great tutorials! keep it coming!
Sure, to do it in the third way you have to download Stable Diffusion UI and then install it as I show in the video. You just drop the image again and thats it. No problem! If you have any other quesiton please let me know 👍
@@MintedWealthOfficial Thanks again! i've got a Mac M1 and i have no idea of how could i install St. Diffusion UI on my computer. But thanks anyway! I'll keep watching ur vids. If some day you can do a tutorial to install SD UI in Mac M1 you'l make me happy! :) Thanks again!
I dont have any Mac to make a tutorial. Stable Diffusion UI its also called Automatic 1111, here you can find instructions to install it on Mac github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Installation-on-Apple-Silicon
sad talker is always busy
Thanks for watching! Yes SadTalker is always busy, thats why I recommend to use google colab or your own instalation of Stable Diffusion UI, those are better alternatives. 👍
@@MintedWealthOfficial can u do a tutorial for it, i always use the colab and it just gives me an error message : the result folder not found or sum like that
@@The_creepy_one1 Sure I will do a tutorial on how to install it locally
Great Video but Foocus looks nothing like in the video. Maybe I missed something
For the animation I used Stable Diffusion UI with SadTalker 👍
I think creators should focus more on bringing value than creating talking characters...but whatever works for you😀
I agree with you, the most important thing is bringing value, but if you want to have an animated character in your videos, this is a Great idea and it's also a way to differentiate from others 👍
Sad talker never works for me. That's sad.
Is there anything I can help you with?
It took me more than an hour. To generate it, it was in queue, and after that, it started processing.
And after 30 minutes of processing..... Error.....
Yeah, thats the thing with hugging face, I recommend you to use the last option that iis to install Stable Diffusion UI and then SadTalker, its what I use all the time. 👍
Huh... Give it another year, and there will be no real people on the Internet. Just AI avatars.
Thank you for watching, yes there are many tools now for creating AI avatars
@@MintedWealthOfficial - you wanna hear something wild? I just had a "conversation" with the Gemini AI, about a story I started 35 years ago.
The story involves synthetic, engineered people. Some of which start out as very advanced AIs.
I sure as hell never imagined that happening, when I started writing it in 1986. LMAO
( That was also the year I first sat down and worked on a computer. )
Looks like your story was ahead of its time - you were predicting the future without even realizing it! Thank you for sharing 👍
mmmmmmmmmm.........?
?
Tried with google colab but still not working.
Made sure the image is png and audio is mp3 named Vocals.mp3.
When I run, it says:
FileNotFoundError: [Errno 2] No such file or directory: './examples/driven_audio//content/SadTalker/examples/driven_audio//content/SadTalker/examples/driven_audio/Vocals.mp3'
I think I found the error:
Audio should just be /examples/driven_audio/Vocals.mp3
Yes that's correct, you had the path twice. I always recommend the last option, it's more reliable and it Is not hardware intensive. 👍
@@MintedWealthOfficial The animation ran but when it was complete there was no .mp4 file in the results folder. By the last option do you mean the stable diffusion web UI? Do I need to install this on my device to access the screen in your video at 6:46 ?