Feature
AI avatar video generator: one photo becomes a talking presenter
An AI avatar video shows a presenter talking without anyone being filmed. Cutroom makes one from a single photo and an audio track, then edits it into a vertical ad.
In short
- An AI avatar video generator makes a video of a talking presenter without filming anyone.
- Cutroom has no ready-made avatars. You bring one photo, and Cutroom makes that face talk along with your audio.
- You can record your own voice for free, or type a script and Cutroom reads it out for 70 credits a minute.
- HeyGen and Synthesia are better if you need ready-made avatars, many languages or long training videos.
- Cutroom's avatars come with Basic at $39.99 a month, which covers about seven 40-second avatar ads with your own voice.
- Photo needed
- 1
- Basic a month, the cheapest plan with avatars
- $39.99
- 40-second avatar ads a month on Basic
- ~7
What is an AI avatar video generator?
It makes a video of a person talking from a photo or a ready-made avatar, plus a script or an audio track. Nobody has to stand in front of a camera.
- Most tools, like HeyGen and Synthesia, start from a script and a ready-made avatar or a digital copy of you.
- Cutroom starts from one photo you supply and an audio track, and animates the mouth, jaw and head.
- People use avatar videos for ads, training and product explainers when nobody wants to be on camera.
How Cutroom makes an avatar video
You upload one photo and a voice recording, or type a script instead. Cutroom makes the face talk, then edits the result like a video you filmed yourself.
- You upload one photo of a face looking at the camera.
- You upload a voice recording, or type a script and Cutroom reads it out for 70 credits a minute.
- Making the avatar takes a while. For 30 seconds of audio it usually takes under 20 minutes, and Cutroom emails you when it is ready.
- The editor then adds captions and short clips of what the presenter talks about. It uses your own product clips if you upload them, and free stock clips if you do not.
From one photo to a finished ad
You can download the avatar video on its own, or send it straight to the editor.
One photo
Facing the camera, mouth clear
Your voice
Your own recording, or a typed script read out
Lip sync
Mouth, jaw and head follow the audio
The avatar video
350 credits per minute
Edit and export
Captions, short clips, 9:16 MP4
This is the whole editor
Highlight a phrase and a clip lands on those exact words. No timeline, no keyframes, no layers.
What the finished avatar video looks like
The face from your photo talks in time with your audio. Only the mouth, jaw and head move.
- The clothes, the background and the light stay as they are in the photo.
- The avatar has no hands, so it cannot hold or point at your product. To show the product, upload your own clips of it, and Cutroom plays them on the words that name it.
- It does not move its body or look away, and it keeps the same expression from start to end, so short videos work best.
- Some viewers can tell it is an avatar, so for health or money claims a real person on camera is more convincing.
- After the edit you get a vertical 9:16 MP4 with captions and short clips over the key words.
A finished Cutroom ad
This example is not an avatar. An avatar ad gets the same headline, captions and clips, around the face from your photo.
What you need: one photo and one audio track
A sharp, front-facing photo gives a sharper video, and clear audio gives better lip sync.
- Use a sharp photo of the head and shoulders, not a full-body photo, so the face fills most of the picture.
- You can use your own face, someone who agreed in writing, a stock model whose licence covers AI video, or an AI-generated portrait. Public figures are never allowed.
- A 30-second test costs 175 credits and shows how well your photo works.
What works best
| What | What works best |
|---|---|
| Photo angle | Facing the camera, not turned to the side |
| Expression | A relaxed mouth, closed or slightly open, not a wide grin |
| Light | Soft and even, with no hard shadows on the face |
| Mouth and jaw | Nothing in front of them: no hand, microphone or shadow |
| Audio | Clear speech at an even pace, up to 5 minutes long |
| Background sound | None: no traffic and no music mixed in |
| Room | A small room with curtains or a sofa, so your voice does not echo. Echo makes the lip sync less accurate. |
| Voice | Your own recording (free) or a typed script read out (70 credits a minute) |
What: Photo angle
- What works best
- Facing the camera, not turned to the side
What: Expression
- What works best
- A relaxed mouth, closed or slightly open, not a wide grin
What: Light
- What works best
- Soft and even, with no hard shadows on the face
What: Mouth and jaw
- What works best
- Nothing in front of them: no hand, microphone or shadow
What: Audio
- What works best
- Clear speech at an even pace, up to 5 minutes long
What: Background sound
- What works best
- None: no traffic and no music mixed in
What: Room
- What works best
- A small room with curtains or a sofa, so your voice does not echo. Echo makes the lip sync less accurate.
What: Voice
- What works best
- Your own recording (free) or a typed script read out (70 credits a minute)
What an AI avatar video costs in Cutroom
The avatar is the most expensive step, because every frame is generated.
Avatar prices and plans
| What | Cost |
|---|---|
| Avatar video | 350 credits per minute |
| Generated voice-over | 70 credits per minute |
| Your own recorded voice | Free |
| Edit: cuts, captions, clips | 100 credits |
| Export | 20 credits per minute |
| A 30-second test | 175 credits |
| A 40-second avatar | About 233 credits |
| A 40-second avatar ad with your own voice | About 347 credits, edit and export included |
| Basic, $39.99 a month | 2,500 credits, about seven 40-second avatar ads with your own voice |
| Premium, $79.99 a month | 5,000 credits, about 14 of those ads |
| Top-up, $39.99 | 2,500 credits, about 7 more of those ads |
| Free first video and Lite | No avatars |
What: Avatar video
- Cost
- 350 credits per minute
What: Generated voice-over
- Cost
- 70 credits per minute
What: Your own recorded voice
- Cost
- Free
What: Edit: cuts, captions, clips
- Cost
- 100 credits
What: Export
- Cost
- 20 credits per minute
What: A 30-second test
- Cost
- 175 credits
What: A 40-second avatar
- Cost
- About 233 credits
What: A 40-second avatar ad with your own voice
- Cost
- About 347 credits, edit and export included
What: Basic, $39.99 a month
- Cost
- 2,500 credits, about seven 40-second avatar ads with your own voice
What: Premium, $79.99 a month
- Cost
- 5,000 credits, about 14 of those ads
What: Top-up, $39.99
- Cost
- 2,500 credits, about 7 more of those ads
What: Free first video and Lite
- Cost
- No avatars
AI avatar vs filming yourself
Film yourself when people need to trust a real person. Use an avatar when nobody can be on camera.
Filming yourself and an AI avatar side by side
Holding the product
- Filming yourself
- Yes, your hands are in the shot
- An AI avatar
- No hands and no props
Trust on health or money claims
- Filming yourself
- Higher, a real person says it
- An AI avatar
- Lower, some viewers spot an avatar
Someone on camera
- Filming yourself
- Yes, you
- An AI avatar
- No, a photo is enough
Changing the script
- Filming yourself
- Film it again
- An AI avatar
- New audio, same photo
A 30-second ad in Cutroom
- Filming yourself
- About 110 credits
- An AI avatar
- About 285 credits with your own voice
Plans
- Filming yourself
- Every plan, first video free
- An AI avatar
- Basic and Premium
| Filming yourself | An AI avatar | |
|---|---|---|
| Holding the product | Yes, your hands are in the shot | No hands and no props |
| Trust on health or money claims | Higher, a real person says it | Lower, some viewers spot an avatar |
| Someone on camera | Yes, you | No, a photo is enough |
| Changing the script | Film it again | New audio, same photo |
| A 30-second ad in Cutroom | About 110 credits | About 285 credits with your own voice |
| Plans | Every plan, first video free | Basic and Premium |
Which AI avatar video generator should you choose?
HeyGen and Synthesia give you ready-made avatars that read a script in many languages. Cutroom has no ready-made avatars. It makes your photo talk, then edits the result into a vertical ad.
- Choose HeyGen if you want a digital copy of yourself or videos in many languages.
- Choose Synthesia if you make training or internal videos and want ready-made avatars.
- Choose D-ID if you want a cheap talking photo and do not mind a watermark on its Lite plan.
- Choose Cutroom if you want the avatar turned into a finished vertical ad, with captions and short clips of what it says.
AI avatar video generators compared
Prices as each company publishes them, checked 21 September 2026.
| Tool | Best for | Price from |
|---|---|---|
| Cutroom | Vertical ads from one photo and your audio | Basic $39.99/month; 350 credits per avatar minute |
| HeyGen | Avatar videos from a script in 175+ languages | Free plan; Creator $24/month yearly ($29 monthly) |
| Synthesia | Training and internal videos with avatars | Free plan; Starter $18/month yearly ($29 monthly) |
| D-ID | Talking photos and interactive agents | 14-day trial; Lite $4.70/month yearly |
Tool: Cutroom
- Best for
- Vertical ads from one photo and your audio
- Price from
- Basic $39.99/month; 350 credits per avatar minute
Tool: HeyGen
- Best for
- Avatar videos from a script in 175+ languages
- Price from
- Free plan; Creator $24/month yearly ($29 monthly)
Tool: Synthesia
- Best for
- Training and internal videos with avatars
- Price from
- Free plan; Starter $18/month yearly ($29 monthly)
Tool: D-ID
- Best for
- Talking photos and interactive agents
- Price from
- 14-day trial; Lite $4.70/month yearly
Questions people ask
- Can I use my own voice or a typed script?
- Yes, both. Your own recording is free and usually sounds better. You can also type a script, and Cutroom reads it out for 70 credits a minute.
- How much does an AI avatar video cost?
- In Cutroom, 350 credits per minute of avatar video, plus 100 to edit and 20 credits a minute to export. A 40-second avatar ad with your own voice is about 347 credits, so Basic ($39.99 a month) covers about seven and Premium ($79.99) about 14.
- Is there a free AI avatar video generator?
- Yes. HeyGen's free plan makes 3 videos a month of up to 1 minute, with a watermark. Synthesia has a free plan too. Cutroom's avatars need Basic at $39.99 a month.
- Which languages can the avatar speak?
- Any language your audio is in, because the lips follow the sound. You can also type a script in Dutch, for example, and Cutroom reads it out. Cutroom's editor captions an avatar video in English, so for Dutch captions, download the avatar video and upload it again as a Dutch video.
- How long should an avatar video be?
- About 40 seconds for an ad. The avatar for that costs about 233 credits. The face keeps the same expression the whole time, which people notice more in a longer video. The audio can be up to 5 minutes long.
- Can I use a photo of a celebrity or a stock model?
- Not a celebrity: public figures are never allowed. A stock model is fine only if the licence covers AI video. Your own face, someone who agreed in writing and an AI-generated portrait all work.
- Should I say the presenter is AI?
- Yes. Several platforms require a label on realistic AI people, so add one before you post.
Make a talking presenter from one photo and your voice. Avatars come with Basic, $39.99 a month for 2,500 credits.