Impressive, but fundamentally void of humanity and depth. Nothing generated will compare to the level of expertise that real animators and artists put in, including those who had their original work stolen for this concoction to be created. Public AI programs like this do not ask permissions. This is a threat to the income and livelihood of artists. The only AI that should be publicly available is those which concern information, statistics, and deductive reasoning. At a push, art AI should strictly be programmed with royalty-free art to avoid copyright infringement and defilement of existing work. Don’t get me wrong, there ARE artists using AI in very creative and ethical ways. The difference is that these artists are also programmers who have built their own AI programs from the ground up. This means the AI, input, and final output are all original and rightfully belong to them. I really recommend seeing some of Holly Herndon’s work as an example of these creative endeavours. Good luck with your work, though I hope one day you are able to produce viable art with originality. It saddens me to see creative endeavours skewed to this extent for the sake of quick production and cash. Very dystopian.
Wow Zinny, that was amazing. I actually saw a channel recently, that I'm pretty sure the girl learned how to create from you. I have seen your videos before and recognized the style of your format for your AI character. Her video was also pretty good.
Wow, this is disgustingly bad. I'm currently waiting to go to arts uni for an animation course. I don't think I'll ever have to worry about my future jobs being taken by this slop
Subbed and liked. Thanks for the great tips! I was holding off on making AI videos until something like Dream Machine came along. Even with the best prompts it still makes lots of mistakes, but the keepers are worth it.
@@thezinnystudio Have you tried Runway Gen 3? It can’t do image to video yet but using Midjourney prompts to describe the scene works surprisingly well. Plus Runway can do lip syncing as a bonus.
I have been abit skeptical because of the lack of ability to do image to video. I will try the prompt directly and put it to test and see how it holds up. Thanks for bringing my attention to it again. 🙂
Hey Zinny, great tutorial as always. I created a few videos using these tools and am currently working on a video story. I'm finding that 1. Dream Machine does not do well with hands. Sometimes they come. out looking really weird, or like blobs of flesh. (have you come across this and have you found a solution. 2. Let people know even though it's free, the time to render a video is slooooow. I mean sometime a few hours. Also if you are doing a project that requires multiples scenes, you might as well get a paid plan, as you will go through the initial amount they give you very fast. I paid, for a plan with 150 videos on Friday, and I'm almost out of videos. 3. Synclabs is good but I have found it gives me glitches. Have you come across this too, and do you have a remedy. I may be doing something wrong. I do know it is best to make sure your subject is facing the camera so it can pick up the lips and the face movements better. That said, I really love this tool. Would appreciate any feedback or guidance you can give. Thank you
1. Yes, dream machine morphs hands sometimes. I haven't found a solution yet, because the model is barley 3weeks old and we are not given any form of control. Some AI tools like Midjourney that had similar issues initially, but with time they got better and this is the work of the Data scientist that work on this model not us. We will just have to wait it out, for the hand issue to be fixed. 2. I guess i skipped that because i am also using a paid plan, which is way faster that the free plan. 3. For Synclabs, i also noticed the glitching, but in scenarios where, the subject in the video is abit far and moving alot and also the model you are using. What i find to work, is the subject, been close and not move alot and i also tired all the models they have to see which one gives me a better result. The tool, struggles to follow the mouth if the movement is alot. but when the subject is abit steady it gives a better result. Like i said earlier, with time these ai tools will fix their limitations, but before then, we either find a way around it or wait for next models to be released.
@@thezinnystudio Thanks for the feedback. Yes, I agree, I need to just keep working with. it. and I am finding that I. need to insure the subject is facing front with little movement to get the lip sync right. I didn't realize the model was only 3 weeks old. So it makes sense. Well I'm having tons of fun with it creating my childrens channel. If this is the worse it gets, the future will be phenomenal! last thing, besides Pika labs, have you found any other tools that lipsync the video?
.•:* ̈ ̈*:•.WONDERFUL* ̈ ̈*RECOMMEND* ̈ ̈* Great tutorial. Would be great to know how to come up with good suno prompts. Thanks for support. Best Regards
All the tools I mentioned has a free plan expect midjourney. You will have to visit each and see the plan that works for you, then calculate how much it will cost you.
Finally, thanks for the upload, zinny. Thanks to your tutorial, my daughter can create a talking avatar. One quick question: Do we need to create multiple scenarios before we get a perfect fit for lip syncing?
Majority of the time the tools get it in one trial. I noticed that it depends om the source video. If video is not changing too much and is up close. It is better than when there are alot of movement and far off.
At the based of the new video. You will see the button extend, click on it, type your prompt and also add the image, then enter. It will use the information to extend the already existing video.
The video is amazing - but when I upload a song to Pika for lip syncing, I keep getting a message "something went wrong", no matter how often I try. Any idea what the reason could be?
This could be the issue, if your audio is longer than the video clip, and you want to use it that way, it wont work. Rather you have to clip the audio and video to be the same length before syncing. Pika, gives you box, where you can clip your audio to the video length. if the audio is long you will have to repeat the same process, for different timelines of the audio.
@@thezinnystudio After many attempts I figured it out. There is also the limiting factor that the videos should not be bigger than 10MB in size. I tried with Runway as well, there you can do longer clips, but it is soooo expensive. 5 credits per second of Lip Sync. Thank you so much for your tutorials, they are really helpful. My first modest attempt is this (cutting off before the end of the song to fit the short length...still practicing :) ua-cam.com/users/shortsxaQrgreLO1c
LOVE YOUR VIDEOS! 😍 I’m inspired to make my own. But can I ask you a question? What category do you put your videos under? Technology or education or something else? Please help!
I tried to follow you but got stuck on Pika. I heard you say that the video is shorter than the song and that you made clips but I keep getting an error, even if I clip the audio. Can you please show more clearly how to extend the video to match the audio? I'm stuck.
For Pika, if the video you are trying to lipsync is around 20 secs and the voice is like 30secs. When you upload the voice try and clip it to about 18secs to the video and try. It should work
@@thezinnystudio The video is only 5 secs and the song is 2 min. I’m confused since the song would not play out. Am I supposed to build 5 seconds clips at a time. That’s a lot of clips.
Unfortunately yes for Pika art. For tools like sync lab, it does it for you. Since you are already using Pika you will have to break it up or else it won't sync
Thank you for the tutorial. What does it mean by non-commercial use for the free account please? Does it mean you can’t upload to UA-cam and monetise? 😊
It means you cant use the creation for any monetization purposes including UA-cam. if you chose to use for UA-cam content, you will need to get a paid plan.
How did you lipsync the entire song with only a 4 second clip From Luma? Did you loop the 4 seconds repeatedly til it reached the entire length of the song?
Yes. Clip the first 4 seconds to the video. Then upload the video again and shift the trim bar to the next 4seconds, that should be 5seconds to 8seconds and generate video. Follow the same process to get to the end then download all the lipsynced video and then merge into one.
That editing to be honest is pretty bad You basically just put a bunch of videos together and they're not transitioning in between perfectly They do that jump thingy That's why this technology is not that great yet
If you listened to the video I said. One can go ahead to edit the video the way they want. The essence is to show how the combined clips into one video looks like.
I have watched many videos on how to create animations, to be frank, this is one of the best videos I have watched so far. I had to pause the video and comment before I finished the remaining part. Thank you so much for sharing.
You can do that using Midjourney. But in this tutorial, i just use the same image and dream machine, gave me multiple videos with different actions of the same character.
for people who are not poor and want a better set-up just use a metahuman and iphone and unreal engine. Works way better than this and is much faster, this is more of the lower end tutorial
Thank you so much for watching. Do not forget to like the video and subscribe to the channel
Good evening mam please 🙏can African folktales stories be monetize on Facebook?
Thank you so much for your time and information 😊 truly appreciate it...keep up the great work and be blessed
@@thezinnystudio thanks for great videos and responding to comments!
Zinny, you are absolutely the best at explaining what to do and how to do it, and for free (Mind Blown)!! I love all your videos. Keep 'em coming!
Thank you so much!
Impressive, but fundamentally void of humanity and depth. Nothing generated will compare to the level of expertise that real animators and artists put in, including those who had their original work stolen for this concoction to be created. Public AI programs like this do not ask permissions. This is a threat to the income and livelihood of artists. The only AI that should be publicly available is those which concern information, statistics, and deductive reasoning. At a push, art AI should strictly be programmed with royalty-free art to avoid copyright infringement and defilement of existing work.
Don’t get me wrong, there ARE artists using AI in very creative and ethical ways. The difference is that these artists are also programmers who have built their own AI programs from the ground up. This means the AI, input, and final output are all original and rightfully belong to them. I really recommend seeing some of Holly Herndon’s work as an example of these creative endeavours.
Good luck with your work, though I hope one day you are able to produce viable art with originality. It saddens me to see creative endeavours skewed to this extent for the sake of quick production and cash. Very dystopian.
All this plenty text just so you don't have to say, "good job! I am proud of you" You need to heal of the pains in your heart granola
I want to see a video where DallE-3 and other AI image generators advance to where two customized talking avatars can be in one scene.
Noted
sorry the lipsync is terrible
“Synclabs” is way better for lip syncing.
@@naomithemarketerhow much the monthly plan for Synclabs?
Well done and thank you . AI is getting really good
Yes it is
Love it! Yes please do an updated on image creation step by step!
Will do!!
May I please receive an updated version for step by step creations.
Thank you.
Go deliver pizza instead 🤣🤣🤣
thanks for watching the video.
Wow Zinny, that was amazing. I actually saw a channel recently, that I'm pretty sure the girl learned how to create from you. I have seen your videos before and recognized the style of your format for your AI character. Her video was also pretty good.
Thanks. I teach alot about my process, I am happy even if one person starts their own thing, from what they have learnt.
@@thezinnystudio I know Zinny. I had to let you know and basically give you your flowers 💐
Thank you 🙂
Wow, this is disgustingly bad. I'm currently waiting to go to arts uni for an animation course. I don't think I'll ever have to worry about my future jobs being taken by this slop
Tbh the lip sync is only ok, but a video about ALL prompts would be great!
How about you get a real talent
Thanks for watching
The teeth 😨
Hi Zinny!
Could you please explain what are the rules for a AI channel in youtube to get monetization
Yes, please show me image generation on Midjourney.
I struggle with making my characters, please create video(s) on image creation.
I struggle with making my characters, please create a video(s) for image creation.
Very nice, wish there was one app that did it all and supported long format videos at a resonable price
Thank ypu, and yes please do an update on image creation
Alright noted
Yes, +1 for this, thanks!
Subbed and liked. Thanks for the great tips! I was holding off on making AI videos until something like Dream Machine came along. Even with the best prompts it still makes lots of mistakes, but the keepers are worth it.
for sure. just note that it is only going to get better from here
@@thezinnystudio Have you tried Runway Gen 3? It can’t do image to video yet but using Midjourney prompts to describe the scene works surprisingly well. Plus Runway can do lip syncing as a bonus.
I have been abit skeptical because of the lack of ability to do image to video. I will try the prompt directly and put it to test and see how it holds up. Thanks for bringing my attention to it again. 🙂
you have a very calming and beutiful voice
Hey Zinny, great tutorial as always. I created a few videos using these tools and am currently working on a video story. I'm finding that
1. Dream Machine does not do well with hands. Sometimes they come. out looking really weird, or like blobs of flesh. (have you come across this and have you found a solution.
2. Let people know even though it's free, the time to render a video is slooooow. I mean sometime a few hours. Also if you are doing a project that requires multiples scenes, you might as well get a paid plan, as you will go through the initial amount they give you very fast. I paid, for a plan with 150 videos on Friday, and I'm almost out of videos.
3. Synclabs is good but I have found it gives me glitches. Have you come across this too, and do you have a remedy. I may be doing something wrong. I do know it is best to make sure your subject is facing the camera so it can pick up the lips and the face movements better.
That said, I really love this tool. Would appreciate any feedback or guidance you can give.
Thank you
1. Yes, dream machine morphs hands sometimes. I haven't found a solution yet, because the model is barley 3weeks old and we are not given any form of control. Some AI tools like Midjourney that had similar issues initially, but with time they got better and this is the work of the Data scientist that work on this model not us. We will just have to wait it out, for the hand issue to be fixed.
2. I guess i skipped that because i am also using a paid plan, which is way faster that the free plan.
3. For Synclabs, i also noticed the glitching, but in scenarios where, the subject in the video is abit far and moving alot and also the model you are using. What i find to work, is the subject, been close and not move alot and i also tired all the models they have to see which one gives me a better result. The tool, struggles to follow the mouth if the movement is alot. but when the subject is abit steady it gives a better result. Like i said earlier, with time these ai tools will fix their limitations, but before then, we either find a way around it or wait for next models to be released.
@@thezinnystudio Thanks for the feedback. Yes, I agree, I need to just keep working with. it. and I am finding that I. need to insure the subject is facing front with little movement to get the lip sync right.
I didn't realize the model was only 3 weeks old. So it makes sense.
Well I'm having tons of fun with it creating my childrens channel. If this is the worse it gets, the future will be phenomenal!
last thing, besides Pika labs, have you found any other tools that lipsync the video?
This was so cool! Nice work, subbed and liked!
Thanks
Love you so much! ❤
Thank you for your love and support!
what a clear and straightforward tutorial...thanks for this! Subscribed!
Uwc, welcome to the community
.•:* ̈ ̈*:•.WONDERFUL* ̈ ̈*RECOMMEND* ̈ ̈* Great tutorial.
Would be great to know how to come up with good suno prompts.
Thanks for support. Best Regards
use chatgpt, it will give you interesting suggesions
How much does all this cost?
All the tools I mentioned has a free plan expect midjourney. You will have to visit each and see the plan that works for you, then calculate how much it will cost you.
Thankyou for this video. Inspired me to start making content ❤❤❤
Go for it!
Finally, thanks for the upload, zinny. Thanks to your tutorial, my daughter can create a talking avatar. One quick question: Do we need to create multiple scenarios before we get a perfect fit for lip syncing?
Majority of the time the tools get it in one trial. I noticed that it depends om the source video. If video is not changing too much and is up close. It is better than when there are alot of movement and far off.
Thank you so much for just teaching what you learn. Please can you do a video on how to generate images on midjourney.
Noted
Hi Zinny Question does Runway ML do the same as Luma Ai, meaning you could create the same singing avator in Runway ML?
Yes, using the gen 3 alpha you can make the and animation sing then use other tools I talked about to do the lipsync
Thanks Zinny! Great tutorial!
Glad you like it!
Wow you are wonderful dear. Just wished i could subscribe again and again
Thank you so much for your kind words! I truly appreciate your support.
HI good morning, can you show us how to get our videos / content on the kids channel? Thank you so very much for helpinfg us grow our channels.
Noted
How to extend the video to more than 4 seconds?
At the based of the new video. You will see the button extend, click on it, type your prompt and also add the image, then enter. It will use the information to extend the already existing video.
The video is amazing - but when I upload a song to Pika for lip syncing, I keep getting a message "something went wrong", no matter how often I try. Any idea what the reason could be?
This could be the issue, if your audio is longer than the video clip, and you want to use it that way, it wont work. Rather you have to clip the audio and video to be the same length before syncing. Pika, gives you box, where you can clip your audio to the video length. if the audio is long you will have to repeat the same process, for different timelines of the audio.
@@thezinnystudio After many attempts I figured it out. There is also the limiting factor that the videos should not be bigger than 10MB in size. I tried with Runway as well, there you can do longer clips, but it is soooo expensive. 5 credits per second of Lip Sync. Thank you so much for your tutorials, they are really helpful. My first modest attempt is this (cutting off before the end of the song to fit the short length...still practicing :) ua-cam.com/users/shortsxaQrgreLO1c
yeah i'm on this already too, your video is ABSOLTELY AWESOME :)
Thank you
If you have a course send link I want to enroll
Not yet soon
BTW, incredible song and video!
Thanks
I love it!
hi
Hi pls is everything you used free version or paid version?
All the tools has a free plan, when you exhaust your freebies, you will be required to pay for a plan.
yes make image genaration video in mid journey
Noted
Its weird how it jitters randomly, also its a piano not a guitar in the song lol😊
"just got easier" (cuts to different shots with no cohesion or reason
Actually Wapoo is similar in anime animations!
Maybe you wanna have a try!
Very interesting but 8$ + 24$ + 9$ = 41$/month
Your audio is broken for the right ear when using earphones
can you please create an updated tutorial to generate these images in Midjourney?
on it
Am i the only one that can't access luma labs in the whole universe 😢
Thanks alot for this am going with this
UWC
How did you create Letter Z for your Logo ?
I used Midjourney
Hi Again Zinny, lip syncing for audio is it best to use DiD or Sync Labs?
Both are for different purposes. D-ID is for image lipsync while synclab is for video lipsync. So it depends on what you want to do
@@thezinnystudio Thankyou understand now. X
Please which software did you use for the singing avatar? Luma isn't responding at my end.
Kindly check the description of the video all the tools I used are mentioned
How do you make a tutorial like that?
What type of tutorial are you referring to
LOVE YOUR VIDEOS! 😍 I’m inspired to make my own. But can I ask you a question? What category do you put your videos under? Technology or education or something else? Please help!
Education
More on prompt generation
Noted
Lumalabs is down 😮
New Sub. Awesome Content Zinny. Maybe I can make travel animations ❤
Absolutely. The sky is literally your starting point.
I've got a question, can we monetize our video with AI voice? I've been seeing a lot of videos where they say you can't.
Yes you can monetize ai voice. When choosing one, make sure it sounds realistic and doesn't sound like a robot.
Okay but it must be used for good
I tried to follow you but got stuck on Pika. I heard you say that the video is shorter than the song and that you made clips but I keep getting an error, even if I clip the audio. Can you please show more clearly how to extend the video to match the audio? I'm stuck.
For Pika, if the video you are trying to lipsync is around 20 secs and the voice is like 30secs. When you upload the voice try and clip it to about 18secs to the video and try. It should work
@@thezinnystudio The video is only 5 secs and the song is 2 min. I’m confused since the song would not play out. Am I supposed to build 5 seconds clips at a time. That’s a lot of clips.
Unfortunately yes for Pika art. For tools like sync lab, it does it for you. Since you are already using Pika you will have to break it up or else it won't sync
@@thezinnystudio okay thanks for your help!
OMG!! I fell in love ❤️
Thanks
Thank you for the tutorial. What does it mean by non-commercial use for the free account please? Does it mean you can’t upload to UA-cam and monetise? 😊
It means you cant use the creation for any monetization purposes including UA-cam. if you chose to use for UA-cam content, you will need to get a paid plan.
Lovely video.
Thank you very much!
Your tutorial part went mute.
your so Amazing my dear... how do you come up with the story for video clip or short movies and directing can you please make a video that helps us ❤❤
I will look into that
Amazing content as always Zinny!! please can we have a tuto on consistent image génération using Leonardo AI?
Noted
I love your videos and would definately love a more in depth guide into prompt generation.
Thank you. Noted
How did you lipsync the entire song with only a 4 second clip From Luma? Did you loop the 4 seconds repeatedly til it reached the entire length of the song?
Yes. Clip the first 4 seconds to the video. Then upload the video again and shift the trim bar to the next 4seconds, that should be 5seconds to 8seconds and generate video. Follow the same process to get to the end then download all the lipsynced video and then merge into one.
@@thezinnystudio thank you so much! 🙏
The fingers were amazing
Yooo
great vid
Thanks for watching, glad you enjoyed it!
❤❤❤❤❤❤❤❤❤❤❤
Thanks
the teeth bro
That editing to be honest is pretty bad You basically just put a bunch of videos together and they're not transitioning in between perfectly They do that jump thingy That's why this technology is not that great yet
If you listened to the video I said. One can go ahead to edit the video the way they want. The essence is to show how the combined clips into one video looks like.
I have watched many videos on how to create animations, to be frank, this is one of the best videos I have watched so far. I had to pause the video and comment before I finished the remaining part. Thank you so much for sharing.
You're very welcome!
the song is a hit .. seriously i cant stop hearing it ... give us the full version please
Sure
thank you so much for this
You're very welcome!
Thank you so much! i follow all this websites but i had a rough time streamlining the flow. You are awesome!
Thank you. when you say you are having issues streamlining the flow can you explain further, let me see if i can clarify.
Thank you. Please do a video how did you generate the image in midjourney I want to generate the same style for my children's book ❤
Noted
@@thezinnystudio thank you so much for your efforts!
😮😮 am speechless, This is amazing 🤯🤯
Thank you so much for your kind words!
How long does it take to generate a video? Would it be faster if I pay for it?
It is faster if you pay less than 3mins, depend on the image you are animating
@@thezinnystudio thanks so much. ❤️
Looking forward for your next video!
@@thezinnystudiohow to keep my original quality of my video in Pika? Because when I put my video in there, it’s so ugly. Please tell me why.
What exactly is ugly the lipsync or did pika change the video?
@@thezinnystudio the color of the video has changed when I put it in Pika. 🥹
WOW!!!!! Fantastic video!!!! Thank you!
Uwc
How to get multiple images of same avatar
You can do that using Midjourney. But in this tutorial, i just use the same image and dream machine, gave me multiple videos with different actions of the same character.
@@thezinnystudio Thanks
Wow
Amazing 🤩
Thank you so much
I Love your videos!
Thanks
Awesome!!
Thanks!
Copy Paste (F)art.
nice
Thanks
wow
I know right.
Hello thank you I really like the first site, can I publish this music on UA-cam without having problems with the rights of others?
for people who are not poor and want a better set-up just use a metahuman and iphone and unreal engine. Works way better than this and is much faster, this is more of the lower end tutorial
y not show ur face? u shud be proud of how u look like, we all, love ur self
See that up there, yeah that's sense flying away from you my good man 😊
Thanks!
Thank you so much
I know this is about animation but Suno is no joke...how fast it can produce a banger is crazy! Love your stuff sis! You're one of my faves👌🏽🫶🏽
Thank you so much!!