Originally Posted by floyd jane
B.D.

Awesome.

Guitars sound Great.
Bass sounds Great.
Drums sound Great.
Vocals sound Great.

LOVE the melody.
Love the "space" in both the write and the production.

Gorgeous mix.

Love the write-up, too....


Please tell us what you use to do the video. (sites and/or software).
Amazing consistent character. Amazing lip-sync.
Details. please. How much time does it take?

Color me Impressed.
On so many levels.

fj
Thanks a lot for your really nice comment, Floyd.
Highly appreciated smile

The making of the video in a nutshell:
  • It all runs locally on my PC, I hate subscriptions
  • You need an NVIDIA graphics card with at least 16 GB of VRAM, the more the better
  • I use 2 AI models:
    Qwen-Image-Edit to create and modify images
    LTX-2 to create the videos
  • There's a (browser-based) tool called ComfyUI (see https://www.comfy.org/ ) to run the AI models
  • There's a guy on YouTube who creates workflows for using the AI models in ComfyUI: Nerdy Rodent (YouTube) where I learned the basics.
    Everything is now also described on the model pages linked above, but it is still a good idea to watch the videos.


In terms of time required, it took me about a week (2-3 hours per day) to explore all the possibilities and customize the ComfyUI workflow to my needs.
However, you only need to do this once and can then use it for all your videos.
My main focus was on a) character consistency, b) perfect lip synchronization, and c) natural acting and breathing.
For the next video, I plan to make the characters look more natural, but that shouldn't be too difficult.

The final video took about four hours with a lot of trial and error until all the scenes matched my vision.
It's a good idea to write a film script beforehand!
Then another hour to sync the scenes to the music in my video software.

Goldmania Bella
Listen on Spotify
Watch on YouTube
Buy on Amazon