AI Video Glossary: Key Terms Defined
A plain-English AI video glossary defining avatars, digital twins, voice cloning, C2PA, deepfakes, TTS, lip sync, and more key AI video terms.
AI Video Glossary
This AI video glossary tells you what each word means. It covers avatars, twins, voice cloning, and more. Each one is short and plain. You can quote it or share it.
On this page
- AI avatar
- AI disclosure
- AI dubbing
- Avatar consent / likeness rights
- B-roll
- C2PA / Content Credentials
- Deepfake
- Digital twin
- Generative video
- Lip sync
- Localization
- Render
- Stock avatar
- Talking-head video
- Text-to-speech (TTS)
- Text-to-video / script-to-video
- UGC (user-generated content)
- Voice cloning
- Watermark
AI avatar
An AI avatar is a fake one who talks on the screen. No real one is on film, and the app moves the mouth.
AI disclosure
AI disclosure means you tell folks that AI made the clip. It can be a tag, a note, or hid in the file.
AI dubbing
AI dubbing swaps the sound of a clip for a new tongue. It can keep your own voice and move the mouth to fit.
Avatar consent / likeness rights
Consent is the OK you give for an avatar to use your face. Your rights say who can use your face and voice.
B-roll
B-roll is more film laid on top of the main shot. It shows what the one speaks of, and the voice goes on.
C2PA / Content Credentials
C2PA is an open way to show where a file came from. It is like a food tag, and no one can fake it.
Deepfake
A deepfake is a fake clip made by AI of a real one. It shows them say or do what they did not, and looks real.
Digital twin
A digital twin is a copy of one real one made by AI. It is built once from their face and voice, with their OK.
Generative video
Generative video is a clip an AI tool builds on its own. There is no film shot, and it can start from just text.
Lip sync
Lip sync means the mouth moves to fit the words. AI lines up the lips with the sound so it looks real.
Localization
Localization fits a clip to one tongue or place. It can swap the words, dub the sound, and change the text.
Render
To render is to build the last file from all its parts. You set up the work, hit go, and wait for the file.
Stock avatar
A stock avatar is a built-for-all AI one you can pick. It is not just yours, so lots of folks use the same one.
Talking-head video
A talking-head clip shows one who speaks right to the lens. The shot is from the chest up, and it builds trust fast.
Text-to-speech (TTS)
Text-to-speech turns text you type into sound you hear. It uses a made voice, and new ones sound quite real.
Text-to-video / script-to-video
Text-to-video turns the words you type into a whole clip. You give it text, and it makes the voice and the shots.
UGC (user-generated content)
UGC is stuff made by plain folks, not a paid pro team. It feels more real, like one who talks to a phone.
Voice cloning
Voice cloning uses AI to copy the voice of one real one. It can then read new text in that same voice and tone.
Watermark
A watermark is a mark put on a clip you may or may not see. It can show who made it, or that AI did the work.
Frequently asked questions
What is AI video?
AI video is a clip that AI makes or changes a lot. It is not all shot on film, and it can mix a few ways.
What is the difference between an AI avatar and a digital twin?
An AI avatar is any AI one on the screen, real or made up. A twin is one kind, built to be one true real one.
Is a consent-based avatar the same as a deepfake?
No. A deepfake shows a real one with no OK to fool you. An avatar with consent has the OK of the one it shows.
How can viewers tell a video was made with AI?
You can look for tags, words on screen, or spoken notes. The file may hold marks that a check tool can read.
Related reading
More from Kyndrify
Make your first video without filming.
Say what you need and the studio makes it: video, images, voices. Start free, no credit聽card.