This is step by step how to generate the most realistic AI videos with the new Google Omni video model.

@ericdoesecom

74.1K followers · usually 4.9K views a post

Views

24.9K

Likes

897

Comments

18

Shares

144

Selling · SaaS › AI Tool

Google Omniseen in 6 videos →

Why it worked

The video provides a step-by-step tutorial on how to use a new and exciting AI tool (Google Omni) to create realistic videos, which appeals to creators looking for efficient and advanced content generation methods. The offer of a custom skill via DM also incentivizes engagement.

Summary

The video explains how to use Google Omni to create realistic AI-generated videos. It details the process of using reference images and audio clips to maintain character consistency and offers a way to obtain a custom skill for creating realistic AI actors.

Structure

1Introduction to Google Omni for AI video generation
2Explanation of Omni's features (consistency, references, editing)
3Creating a reference image for the AI actor using ChatGPT
4Offering a custom skill for realistic AI actors via DM
5Step-by-step process for using Google Omni with actor image, audio, and scene description
6Demonstration of consistent character and voice across multiple generated clips

Original caption

The next generation of AI video models is here #ecommerce #aivideo #googleomni #geminiomni #aiugc

More from @ericdoesecom

Their other posts in the index, biggest breakout first.

7Concluding thought on the ease of AI video creation

Product placement

Google Omni is demonstrated as an AI video generation tool. The creator also mentions using ChatGPT to create an AI actor image and a custom Claude skill.

Call to action

If you want that, just DM me on Instagram @ericdoescom and I'll send it over so that you can have more realistic actors.

On-screen text

How To Create The Most Realistic AI Videos With Google Omni This is step by step how to generate the most realistic AI videos with the new Google Omni video model. Google just announced this new Gemini Omni video model the other day, and it looks like this is the beginning of the next generation of AI video models that makes it a lot easier to get an accurate output of what you're actually trying to make. It has locked character consistency between video generations, It has the ability to take in audio and video as references in addition to images, and it has the ability to edit your videos with reprompting without actually regenerating the entire video. Anyway to create this AI UGC video, I started by creating an image of my AI actor in ChatGPT. This reference image is gonna be what Google Omni uses to make sure that my character stays consistent for the entire video length through all the generations. To make an image of an AI creator that's tuned for realism and doesn't have that plastic AI face I use a custom Claude skill that I made. If you want that, just DM me on Instagram @ericdoescom and I'll send it over so that you can have more realistic actors. Next, you wanna go over to Google Omni and attach an image of your actor, an audio clip of the voice you wanna use, and a description of your scene and let it go to work. If you wanna tweak the output, just re prompt and Google will edit it without regenerating the entire video, and if you wanna add another scene, Google will keep your characters and audio completely consistent. For example, here's a second clip I generated for the same video, and the character and voice are exactly the same as the first one. We're getting to the point where creating AI videos is gonna be just as easy as recording it yourself.

Transcript

This is step by step how to generate the most realistic AI videos with the new Google Omni video model. Google just announced this new Gemini Omni video model, the other day and it looks like this is the beginning of the next generation of AI video models that makes it a lot easier to get an accurate output of what you're actually trying to make. It has locked character consistency between video generations. It has the ability to take in audio and video as references and addition to images and it has the ability to edit your videos with reprogramming without actually re-generating the entire video. Anyway, to create this AI U G C video. I started by creating an image of my AI actor in ChatGPT. This reference image is going to be what Google Omni uses to make sure that my character stays consistent for the entire video length through all the generations. Make an image of an AI creator that's tuned for realism and doesn't have that plastic AI face. I use a custom cloud skill that I made. If you want that, just DM me on Instagram@eric.com and I'll send it over so that you can have more realistic actors. Next, you want to go over to Google Omni and attach an image of your actor, an audio clip of the voice you want to use in the description your scene and let it go to work. If you want to tweet the output, just reprint and Google will edit it without regenerating the entire video. And if you want to add another scene, Google will keep your characters in audio completely consistent. For example, here's a second clip I generated for the same video and the character and voice are exactly the same as the first one. We're getting to the point where creating AI videos is going to be just as easy as recording it yourself.