1/6
Hope everyone is having a great weekend! As promised, here's a rundown on how I put together an unofficial video for @Apashe_Music 's "Bad News". I'll keep this thread short to get you hitting the ground and creating and then include more detailed examples at the end. Cool?… twitter.com/i/web/status/1… @Apashe_Music 2/6
Intro
Let's start with the facts. This video is 100% generated in Runway gen2 using text prompts and edited together using Davinci Resolve. No other tools were used in this production. While Resolve is free to use, gen2 is not, and the ~ 100 clips generated to make this video… https://t.co/boMl2QnuQ6twitter.com/i/web/status/1…

Comparing text2video rendering tools.

The Players:
Modelscope v1 - Original A1111 local gens
Videocrafter - A1111 local gens w/ VC model
Runway v2 - Online defined length gens
Zeroscope v2 - Newly released A1111 local gens

The Prompt:
a balloon monster walking on the beach Modelscope v1 - So cute! This is actually a good render that flows well. I may have also gotten lucky on my first render, but this was among my favs even before Zero.

1/10
I have something exciting to share this evening. Have you tried using ChatGPT to enhance your imaginative art? It's not an effective tool for every use case but we should always keep it in our toolbox!

I'm going to share my own prompt with you tonight along with how to use… twitter.com/i/web/status/1… 2/10
First, this is the vanilla Midjourney output, which does show what we were asking for.. as in, it gave him a fancy suit and made him look kind of sad. There's not much emotion, though. There's also that Subway lighting and artifacting (double face, etc.) 🤢

It's dictation time! I decided to utilize #VoiceAttack for #Midjourney prompt dictation and it works quite well!! I can even switch to another chat/IM and use Blue Willow, and possibly even browser tools like #PlaygroundAI. Samples from dictated prompts follow...

#AIArtCommunity… twitter.com/i/web/status/1… Midjourneyv5