A talking AI avatar of you, from one selfie.
To make an AI avatar of yourself that talks, upload one selfie and record at least 10 seconds of your voice. VIDEO AI ME turns the selfie into a front-facing facecam actor, clones your voice, and films the avatar saying your script, lip-synced, in clips of up to 15 seconds that join into one continuous video.
Last updated:
Your face · your voice
one photo · 10 s of voice
Talking samples are on the way. Tap a voice to hear it now.
How do you make a talking AI avatar of yourself?
Three inputs, once: a photo, a voice sample and a script. The photo and the voice are saved to your account, so the next video only needs a new script.
- 01
Upload one selfie
Upload one clear, front-facing photo. VIDEO AI ME turns it into your own AI actor and generates a front facecam view of you, plus side views if you want a video that cuts between 2 or 3 angles.
- 02
Record 10 seconds of your voice
Record in the browser or upload at least 10 seconds of clear speech, up to 2 minutes. Your voice is cloned once and reused for every video, and each script is spoken with ElevenLabs v4.
- 03
Paste your script and confirm
The script is split into parts of at most 15 seconds of speech, and you see the parts, the estimated length and the credit quote before anything is filmed. Each part becomes one lip-synced clip, and the clips are joined into one video.
More on each input: AI avatars and AI voice cloning.
How can a talking avatar video run longer than 15 seconds?
Lip-sync video models film at most 15 seconds of speech at a time. VIDEO AI ME splits your script into parts that fit, films each part as one clip starting on the last frame of the clip before it, and joins the clips, so one video can hold up to 30 clips.
Hi, I'm Sam. Everything you see here started from one selfie and ten seconds of my voice. No camera, no studio and no second take, just a script I wrote this morning. That script was cut into short parts. Each part is filmed as one clip of fifteen seconds or less, and each clip starts on the exact frame where the one before it ended. Then the clips are joined in order, so the whole thing plays like one take. Want a longer video? Write a longer script, up to the length your plan allows.
- Clip 010 to 13 s
Starts from your facecam view
- Clip 0213 to 26 s
Starts from the last frame of clip 01
- Clip 0326 to 38 s
Starts from the last frame of clip 02
One video · 38 s · one continuous takeEach clip ≤ 15 s of speech
Facecam or multi-plan: which one should you pick?
| Compared on | Facecam | Multi-plan |
|---|---|---|
| Views of you | One front facecam view | 2 or 3 views of the same person, such as front and side |
| Where each clip starts | On the last frame of the clip before it | On the next view, in turn |
| How it plays | One continuous take | Cuts between camera angles, like a two-camera interview |
| Filming order | One clip after another | All clips at the same time |
What can you use a talking AI avatar of yourself for?
Anything you would say to a camera, once the script is written. The avatar is you, so the videos stay on brand without a shoot.
Founder and team updates
Turn a written update into a video of you saying it, for LinkedIn, an investor email or an all-hands recap, without setting up a camera.
Product explainers and onboarding
Walk new users through a feature with your own face on screen, in 16:9 for the help center or 9:16 for social.
Talking-head ads in your voice
Make UGC-style ads with your face and your voice for Reels, TikTok and Shorts, and test a new hook by changing only the script.
Course and training lessons
Turn lesson scripts into talking-head lessons that run for minutes, not seconds, filmed in short clips and joined.
Sales and customer follow-ups
Answer a prospect or a customer with a short video of you, written instead of recorded.
Multi-angle videos
Use 2 or 3 views of yourself and let the video cut between them, so a long talk does not sit on one frame.
See finished videos on the examples page.
How many avatars, voices and minutes does each plan include?
Each custom AI actor is one person made from one photo, and each cloned voice is one voice sample. Voice minutes reset every month.
| Plan | Price | Custom AI actors | Cloned voices | Voice minutes / month | Longest talking video |
|---|---|---|---|---|---|
| Starter | $29/month | 1 | 1 | 10 | 3 min |
| Pro | $99/month | 3 | 3 | 100 | 3 min |
| Premium | $199/month | 10 | 10 | 600 | About 6 min |
A talking video is capped by the plan's maximum video length and by 30 clips per video. The video is paid with the plan's monthly credits (1 credit = 1 cent), quoted before you confirm, and its voice counts against the voice minutes. Annual billing is 4 months free versus paying monthly. Full plan details are on the pricing page.
Frequently asked questions
How do I make an AI avatar of myself that talks?
Sign up for VIDEO AI ME, upload one clear selfie and record at least 10 seconds of your voice. The selfie becomes your own AI actor with a front facecam view, and the recording becomes your cloned voice. Then paste a script: it is split into parts of up to 15 seconds, each part is filmed as a lip-synced clip of you speaking, and the clips are joined into one video.
How long can a talking avatar video be?
One talking video holds up to 30 clips of at most 15 seconds of speech each, about 6 minutes, and it cannot run longer than your plan's maximum video length: 3 minutes on Starter, 3 minutes on Pro, 12 minutes on Premium. The voice also counts against your monthly voice minutes. For a longer talk, split the script into two videos.
How much of my voice do I need to record?
At least 10 seconds of clear speech, and up to 2 minutes. Record it in the browser or upload an audio file, in a quiet room, speaking the way you want to sound on camera. The voice is cloned once and saved to your account, so every later video reuses it.
Which AI models make the avatar talk?
The voice is generated with ElevenLabs v4 from your cloned voice. The video is filmed with Wan 3, up to 1080p, or Seedance 2.5, up to 720p: each model takes a still of your avatar and one part of the spoken script and returns a lip-synced clip of up to 15 seconds.
What is the difference between a facecam video and a multi-plan video?
A facecam video uses one front view of you, and every clip starts on the last frame of the clip before it, so the whole video plays as one continuous take. A multi-plan video uses 2 or 3 views of the same person, such as front and side angles, and cuts between them from one clip to the next. Both start from the same selfie.
How much does a talking avatar video cost?
Talking videos are paid with your plan's monthly credits (1 credit = 1 cent), at a rate per second of speech that depends on the model and resolution. Before anything is filmed, VIDEO AI ME shows the split into parts, the estimated length and the credit quote, and you confirm. Clips are charged one by one on the real length of the voice. Plans start at $29/month, or $19/month billed annually.
Is there a free AI avatar generator on VIDEO AI ME?
VIDEO AI ME has no free plan and no free trial. Business accounts can get one free demo video during signup, made from the business details they enter, with no card needed. Plans start at $19/month billed annually ($29 month-to-month) and are billed from the first day; every video comes with a full commercial license, and you can cancel in one click.
Can I make a talking avatar of someone else?
Only with their consent. The VIDEO AI ME terms prohibit generating content that depicts real people without their consent, and you may only upload photos and voice recordings you have the rights to use. For a spokesperson who is not you, use an actor and a voice from the built-in library, or get consent from that person first.
Can I use my avatar videos in ads and client work?
Yes. Every paid plan includes a full commercial license: every video you make is yours, client work included, with no watermarks. You can publish the finished video to connected social accounts from the same workspace, or download it.
Ready to hear your own avatar speak?
One selfie, 10 seconds of your voice and a script. Plans start at $29/month; see the pricing page.