AI Video, AI Actor and Avatars

Glintia has three AI tools for moving pictures and people. All of them run in the background and use AI credits, and each shows its price before you start.

Tool	Use it to	Where to find it
AI Video Generator	Generate a video clip from a description, a picture, two frames or reference images	AI Studio › Video, or Generate AI Video in the Video panel
Talking Character (the AI Actor)	Have an animated character or an AI avatar speak your script	AI Studio › Characters › Talking character
Avatars	Place a lifelike presenter and make it talk with Make it talk	Avatars on the rail, or AI Studio › Characters › Avatars

In AI Builder mode, all three have their own rail buttons: AI Video Generator, Talking Character and Avatars. Avatars is also on the rail in the Animation and Full Editor modes.

AI Video Generator

The AI Video Generator gives you 26 video models from providers such as Google (Veo), Kling, ByteDance (Seedance), MiniMax, Alibaba (Wan) and Luma. Pick how you want to start, describe the shot, and the clip renders while you keep working.

Choose how to start

Under Start from, pick one of four ways in. AI Studio’s video tools open the generator on the matching one.

Start from	What you give it	AI Studio shortcut
Text	A description of the scene	Generate a video
Image	One picture to bring to life	Animate a picture
Frames	A first and a last frame, and the clip moves between them	Start → end frame
Refs	Reference pictures of characters, products or places to keep exactly as they look	Consistent characters

The model list only shows models that can do the way you picked. The defaults are Kling 3.0 Pro for Text and Image, Veo 3.1 Fast for Frames, and Seedance 2.0 for Refs.

Generate a clip

Open the AI Video Generator and choose an option under Start from
Under Model, keep the default or click it to choose another. The picker works like the image one: search, Sort, Best for and Must have filters (Sound, Start & end, References, 1440p+, 15 s+ and Loops), plus each model’s price for a 5-second clip and its typical wait
Add your pictures. For Image, fill Picture. For Frames, fill Start and End. In these slots, click Selected to use the image selected on the canvas, or My files to pick one of your uploads or upload a new one. For Refs, add pictures under References with the pointer button (the selected canvas image) or the + button (your files). The hint shows how many references the model takes
Describe the shot. For Text, Describe the video is required: scene, action, camera move and mood. Put dialogue in quotes for models that make sound. For the other options, Describe the motion is optional, for example “slow push-in, hair blowing in the wind”
Pick a Length. Hover over a length to see its price
For Text and Refs, pick a Shape: Slide matches your canvas, or choose an aspect ratio. Image and Frames have no Shape choice, and most models follow the shape of your picture
If the model offers more than one, pick a Quality (resolution)
Set Sound if the model makes it optional
Click Generate video, or press Ctrl+Enter (Cmd+Enter on a Mac) in the description box

Clip settings explained

Length: each model offers its own clip lengths, from 3 seconds up to 30 seconds on the longest models. Veo models make 4, 6 or 8 second clips
Quality: the resolutions the model offers. Higher quality costs more. This is the generated clip’s resolution only. Your export quality still depends on your plan
Sound: models that can make music, effects and speech with the clip show a Sound switch. Some always make sound, and some are silent (add music or a voiceover from the Audio panel)
Seamless loop: on models that support it, the clip ends where it starts, which suits looping backgrounds
+ More options: on models that accept it, a box for things to avoid, such as “blur, text, extra fingers”

The bar above the Generate button shows the total credits, the clip length and quality, and the typical wait.

While it renders

AI video takes from under half a minute to several minutes, depending on the model and length. You don’t need to wait in the panel:

Open My videos to follow progress. Each clip shows Queued or Generating with the time so far and the model’s usual wait
The Background jobs button in the editor’s top bar tracks it too
When it’s ready you get a notice (“AI video is ready”)

When the clip is done, its card in My videos offers Add to timeline, Open (full size in a new tab) and Again (run the same settings again; the button shows the cost). A failed clip says its credits were refunded and offers Retry. Every clip is also saved to your files.

AI video credits

The price depends on the model, length, quality and sound. Some examples at each model’s default quality:

Model	Clip	Credits
Kling 3.0 Pro	5 seconds	70 with sound, 46.75 without
Veo 3.1 Fast	8 seconds at 1080p, with sound	100
Veo 3.1	8 seconds at 1080p, with sound	266.75
Seedance 2.0	5 seconds at 720p, with sound	126.5
MiniMax H3 Max Turbo	5 seconds at 768p	16.75
Grok Imagine Video 1.5 Lite	6 seconds at 720p, from text	15

The cheapest clips start at 5 credits. Long, high-resolution clips on premium models cost far more, so check the total above the Generate button before you start. To draft cheaply, sort the model list by Cheapest first and try the Best value models.

Talking Character (AI Actor)

Talking Character turns a script into a speaking character on your scene. You can use either kind of character:

Animated characters
AI avatars

Glintia’s own characters, marked ANIMATED. The character is animated in the editor with lip-sync and actions matched to what it says. Only the voiceover uses AI credits, and the result stays fully editable. There’s no limit on script length.

Make a character talk

Open AI Studio › Characters › Talking character. The panel walks you through four steps
Script: type what the character says. In the box below it, describe the performance, for example “cheerful, gesturing with hands”. The description shapes how the character performs. Click Next
Character: pick one of the Animated characters or AI avatars. If an avatar has more than one view, choose one under Choose a view. Click Next
Voice: for an animated character, pick a Voice from the list, click Play sample to hear it, or click Browse voices for the full catalogue with filters. An AI avatar comes with its own voice. You can play a sample and tick Generate subtitles. Click Next
Generate: check the summary and click Generate

Progress shows “Generating voiceover…” and then, for an AI avatar, “Rendering video…”. Rendering keeps going if you switch tabs, and you’ll get a notification when it’s ready. The character is added to your current scene. For an animated character, the scene is lengthened if needed to fit the whole voiceover.

Change the script later

With the Talking Character panel open, select the talking character on the canvas. The panel switches to Editing character:

Change the text and click Regenerate. An animated character re-animates, and only the new voiceover uses AI credits.
For an AI avatar, a text-only change reuses the existing video, which is cheaper. Changing the performance description generates a new video.
Click New character to start again from scratch.

Extras for animated characters

After an animated character starts talking, the panel may show two extra sections:

New actions, made for this script: if your character’s actions don’t cover a moment in the script, the AI animator creates a new action for it and places it on the character. These are free and are saved with your custom actions. Click the pencil to fine-tune one. See Custom Actions and Motion Capture
Suggested AI-video beats: moments no action can cover. Click Generate clip to make a short AI video clip (3 to 8 seconds) just for that beat, placed over your character, or Skip it. Each clip uses AI credits: 77.75 for up to 5 seconds, plus 15.5 for each extra second

Video history

The Video history tab lists your recent talking videos with their status (Ready, Generating, Queued or Failed). Click Timeline on a finished video to add it to your design again.

Talking Character credits

Animated character: the voiceover only, at the Text-to-Speech rates (standard voices 2 credits per minute, premium voices 0.5 credits per second).
AI avatar: 0.5 credits per second for the voiceover plus 8.5 credits per second for the video. A 10-second clip costs about 90 credits.

Avatars

The Avatars panel is a library of lifelike presenters, many with several views (for example different poses or settings). Each comes with its own voice. Add one to your scene, write its script, and Glintia makes it talk.

Add an avatar

Click Avatars on the rail (or open AI Studio › Characters › Avatars)
Browse AI Avatars, type in Search avatars…, or narrow the list with the Gender (Male, Female, Neutral) and Location filters. Click the play button on a card to hear the avatar’s voice
Click an avatar. If the card shows “N views”, choose one in the Choose a view window. Views marked Ready to talk render straight away. First use is slower means the view hasn’t been filmed yet: the first script takes about a minute longer, at no extra cost
The avatar is placed on your scene and its Script window opens

Avatars with a lock badge need a higher plan. Clicking one opens an upgrade prompt.

Make it talk

In the Script window, type what the avatar should say. The hint below the box shows the estimated length, for example “≈8s · 20 words”, and tells you if the script is too short or too long. A talking clip must be between 3 and 15 seconds of speech
Check the voice. The avatar opens on the voice it was cast with. To change it, pick another from the Avatar voice list
Click Generate voice to hear the take. This is free. Edit the script and click Regenerate voice until you’re happy
Leave Generate subtitles ticked if you want captions for this clip
Check Estimated cost, then click Make it talk

You can close the window while it renders: it keeps going, and you’ll get a notification when it’s ready. The finished talking video replaces the still picture on your scene, in the same place and size. Avatars that are cut-outs keep their transparent background.

To write a new script later, select the avatar and click Script in its toolbar, then Write the script (or View the result to watch the last render).

Avatar credits

Make it talk costs 5 credits per second of finished video, with the voice and lip-sync included. A 10-second clip costs about 50 credits. A few views can only be animated from the still picture and cost 8.5 credits per second. The Script window always shows the estimate before you render. If you don’t have enough credits for the render, it says “Not enough credits” with your balance, and Generate voice and Make it talk stay disabled.

Troubleshooting

Problem	What to do
”Write a bit more — the voiceover needs to be at least 3 seconds.”	Lengthen the script. Talking video clips need at least 3 seconds of speech
”Shorten the script — AI avatar videos are capped at 15 seconds”	Cut the script to about 220 characters, or split it across two avatars or scenes
”You have reached your AI credit limit for this month.”	Wait for your monthly reset or upgrade your plan. See AI Studio and AI Credits
An AI video failed	Its credits are refunded automatically. Click Retry, or try another model
A clip doesn’t show up on the scene	Open My videos or Background jobs and click Add to timeline