Unfortunately the quality of the video generated by Colab is not as clear as the motion portrait video uploaded. Somewhere along the line it loses clarity. Besides that, it is a very good platform to generate talking head videos.
Hi, Thank you for your comment 🌷. Unfortunately it is true, the face in the final generated video is a little blurry. But there is a walk around solution for this. I use the final generated video and the face image i used in the code, and use them as i make a face swapping. It is not a face swapping actually, but I fool the code to enhance the face quality in the video. And the results were great 👍. You can find the face swapping tutorial here ua-cam.com/video/zZZRrQ1IcUs/v-deo.html
Hi, sorry for the late response 🙏. You can use this method to create videos in any length you want. Just to make it clear, the length of the final video is the length of the audio file used, you cant use a 10 seconds video and 5 minutes audio file, you will get a 5 minutes video as a result.
Hi . Wow this is a good alternative for the D-ID . However I've tried improving the quality by face swapping from your account but it is still the same. Kindly help 🙏
Thank you for this detailed tutorial. I do my stuff on a PC, not on my Android. Does Motion Portrait have a desktop app. I went to their website and tried the demo, but could not download the generated video. I had to screen record that video. Otherwise, it was OK. The final result was not as good as yours.
Thanks for your response. Today, I used a reasonably high resolution portrait photo, the video generated by Motion Portrait was also quite clear, but the resulting final video from Colab was rubbish quality. What do you think went wrong and how to improve the quality of video from Colab? Thanks for some understanding of Google Colab.
Hi, you can create videos up to 5 minutes using this method. The final video duration is the duration of the audio file used If you use an audio file 5 minutes, the final talking video will be 5 minutes.
The duration of the final video is determined according to the duration of the audio file used. You can use an audio file of up to 10 minutes, and the duration of the final video will be 10 minutes. However, the video used must not exceed 60 seconds. You just need to use a video that is only 10 seconds long.
Only you give this valuable work . Which I followed and get success 😊
Thank you so much for your comment 🌷.
I will do my best to keep making useful tutorials for you my friend 🌷.
Thanks 🙏.
It says failed pls do a new video
Unfortunately the quality of the video generated by Colab is not as clear as the motion portrait video uploaded. Somewhere along the line it loses clarity. Besides that, it is a very good platform to generate talking head videos.
Hi, Thank you for your comment 🌷.
Unfortunately it is true, the face in the final generated video is a little blurry. But there is a walk around solution for this. I use the final generated video and the face image i used in the code, and use them as i make a face swapping. It is not a face swapping actually, but I fool the code to enhance the face quality in the video. And the results were great 👍.
You can find the face swapping tutorial here
ua-cam.com/video/zZZRrQ1IcUs/v-deo.html
lo mejor de lo mejor 👍👍
Thanks. The generated video lasts 11sec. How about a video lasting 12 minutes for example? What can you tell us about it? How can we make it?
Hi, sorry for the late response 🙏.
You can use this method to create videos in any length you want.
Just to make it clear, the length of the final video is the length of the audio file used, you cant use a 10 seconds video and 5 minutes audio file, you will get a 5 minutes video as a result.
Hi . Wow this is a good alternative for the D-ID . However I've tried improving the quality by face swapping from your account but it is still the same. Kindly help 🙏
Hi, Thank you for your comment 🌷
Please explain the issue exactly with the face swapping code. Have you got any error messages?
Thank you. The initial process of video without voiceover, does it work for cartoon animal characters.
Hi, Unfortunately in this code, the lip sync doesn't work for animals' faces.
Hi, I try to upload the condensed file in step 2 but get an error message!!
I don't understand step 4, why it always errors
Is there anyway to use eleven labs on Google colab, I want to avoid the character limit
Thanks in advance
Hi, thanks for your comment 🌷.
I will work on "text to speech" soon 👍.
@@BahaaAI looking forward to another awesome tutorial
Please, what is the name of the fourth app you used? I don't really get it. Thanks in advance.
Please, what is the name of the fourth app you used. I don't really get it.
Load photo failed :-/
Thank you for this detailed tutorial. I do my stuff on a PC, not on my Android. Does Motion Portrait have a desktop app. I went to their website and tried the demo, but could not download the generated video. I had to screen record that video. Otherwise, it was OK. The final result was not as good as yours.
Hi, Unfortunately they don't have a PC app.
I use the Motion Portrait app on BlueStacks on my PC and get the same results.
Thanks for your response. Today, I used a reasonably high resolution portrait photo, the video generated by Motion Portrait was also quite clear, but the resulting final video from Colab was rubbish quality. What do you think went wrong and how to improve the quality of video from Colab? Thanks for some understanding of Google Colab.
Dude. You're cool 👍. You got my like & sub.
Any guidance for removing blur from video
Watch this video
ua-cam.com/video/9JCcoSeOmfg/v-deo.html
very nice ...always make to avatar to short video ,your tutorial....thaks brow
Thanks a lot for your comment. Very happy to know that 👍
Very good job in english Mr Baha . I am waiting for the french version !
Thanks a lot 💕
I will try this thank you for the info, i hit sub too. 😊
Thanks 🌷
Thanks sir
But I can't generate voiceover on eleven labs
Hi, Try This : ua-cam.com/video/VzRAMDJoRNY/v-deo.html
Sir I appreciate what you did honestly 😊, I just learned a new thing today.
Thanks 🌷
Step 1: installing wav4lip error..
Fixed, reload the codeand try again, thanks for your feedbak,
@@BahaaAI How to enter the code, I'm not good at it
Just close the project link and reopen it
Thanks sir 🙏 for your excellence teaching this method
It's my pleasure
The best video so far
Thanks 🙏
very good, keep bringing content like this to subscribers
I will my friend 👍
Sir what about 2 or 3 min video how to run in colab plz reply
Hi, you can create videos up to 5 minutes using this method.
The final video duration is the duration of the audio file used
If you use an audio file 5 minutes, the final talking video will be 5 minutes.
@@BahaaAI thanks sir
Sir After Two and Half hours the send error plz any other easy way of talking avatars
Try using another email in colab, and try using colab in incognito tap.
@@BahaaAI okay sir
Step 4 Not working why bro
What is the error message?
@@BahaaAI sorry.. is ok now
thanks brother.... working here.
How to make talking AI bot multilingual like Chat GPT?
Thank you, it's work ..
Thanks 🙏
Thank you
You're welcome 🌷
Loading foto failed, how to fix it
Good Video
Very Good.
Hai sir how to create Cartoon video upload for me sir please iam you tube creater
V good thanks 🙏
Most welcome 🌷
cám ơn anh bạn nhé, tôi muốn gửi cho anh bạn vài ly cafe?
1 minute only ?
The duration of the final video is determined according to the duration of the audio file used. You can use an audio file of up to 10 minutes, and the duration of the final video will be 10 minutes. However, the video used must not exceed 60 seconds. You just need to use a video that is only 10 seconds long.
Why only 10 seconds and not for example 5 minutes? @@BahaaAI
I’m selecting the video but it keep showing error in step 2
😅