The original video demonstrates a 2023 workflow using ChatGPT, Google Bard, text-to-speech, and Adobe’s character animator. Bard is now Gemini and interfaces have changed, but the production logic remains useful. The archive is preserved below.
Play the original tutorial on YouTube. Creator: How To In 5 Minutes.
1. Draft and verify the script
Ask ChatGPT or Gemini for a short scene outline, target age, tone, duration, and visual cues. Treat the output as a draft. Check facts, remove copyrighted characters or imitated living artists, and read the script aloud before generating audio.
2. Create authorized audio
Record your own voice or use a text-to-speech voice licensed for the intended use. Do not clone another person without informed permission. Export clean WAV or MP3 audio and listen for names, numbers, pacing, and awkward pauses.
3. Animate in Adobe Express
Open Adobe Express, choose Quick actions, then Animate characters. Select a character and background, choose the target aspect ratio, and record or upload audio. Adobe generates lip sync plus head, eye, and arm motion. Current recordings can be limited, so split a longer story into scenes.
4. Edit and QA
Preview every scene, trim dead air, add captions, normalize volume, and keep important text inside safe areas. Verify that music, images, voices, and characters are licensed. Export a test and watch it once on a phone before publishing.
Archived companion resources
The original tool roundup remains at Free AI Animation Tools. Other archived links: AI avatar guide, Filmora editor, related creation tutorial, and Kreado AI tutorial.
For continuity, the source URLs remain available: https://youtu.be/7MdI9DmuRLw, https://youtu.be/a7YJqd80joE, and https://youtu.be/IHs7iexV8Rg.
Disclosure
Label synthetic narration or characters when viewers could mistake them for authentic recordings. Never use the workflow for impersonation, fake endorsements, or misleading news footage.
Verdict
This remains a practical beginner workflow when the goal is a simple talking character rather than complex frame-by-frame animation. The strongest results come from a concise verified script and clean audio.