On this page
Short answer
Open Lip Sync on KreatorFlow, add a clear photo with one visible face, then give it a voice: type a script and pick one of our voices, or upload an MP3 or WAV of 4 seconds to 5 minutes. Add an optional direction, choose 720p or 1080p and generate. It costs 4 credits per second of audio at 720p and 6 at 1080p, on any paid plan or credit pack.
Get started
Make a photo talk in five steps
- 1
Choose the photo
Pick a sharp, well-lit photo with one face clearly visible and the mouth uncovered. Head and shoulders or half body works best. In Lip Sync, add it from your device, your uploads, your generations or a saved character.
- 2
Give it a voice
Either type what the photo should say and pick one of our voices (the line is made with Text to Speech, 2 credits per 1,000 characters), or upload your own MP3 or WAV: a recording, a song, or a line made in the Audio studio with a voice you cloned. The track can be 4 seconds to 5 minutes long.
- 3
Direct the performance (optional)
Write one or two sentences about the camera and the motion, such as "Slow push-in. She speaks warmly, nods on the key points and smiles at the end." Do not describe what is already in the photo.
- 4
Pick a resolution
Choose 720p (4 credits per second) for tests and social posts, or 1080p (6 credits per second) for the final version. The cost for your exact track is shown before you generate.
- 5
Generate and download
Press Generate. When the video is ready it is saved to your library, as long as the audio, ready to download and share. If it fails, the credits come back.
What a talking photo costs
Studio credits by length of the voice track, from the same code that charges you.
| Audio length | 720p | 1080p |
|---|---|---|
| 4 seconds | 16 credits | 24 credits |
| 10 seconds | 40 credits | 60 credits |
| 30 seconds | 120 credits | 180 credits |
| 1 minute | 240 credits | 360 credits |
| 5 minutes | 1,200 credits | 1,800 credits |
A 30-second message costs 120 credits at 720p. Lip Sync needs a paid account: any plan (from $15 for the first month on Creator) or a credit pack (from $8). If you use one of our voices, add the Text to Speech price for the line.
How a photo learns to talk
KreatorFlow Lip Sync runs OmniHuman 2.0, ByteDance's audio-driven model, through BytePlus, a ByteDance company (KreatorFlow is an official partner of BytePlus, a ByteDance company). It listens to the voice track and animates the photo to match: mouth shapes for every word, expressions that follow the tone, and natural head and body movement. Because the audio leads, the result looks like the person is actually saying it, not like a mouth pasted onto a still.
It works with photos of people, paintings, cartoon and anime characters, 3D characters and animals. The video above was made from one AI-generated portrait and a typed line in the Dara voice.
Photo tips
The photo decides more of the result than anything else. Before you upload, check it against this list.
- One face, clearly visible, looking at the camera or close to it.
- Mouth and chin uncovered: no hands, microphones, scarves or hair across the lips.
- Head and shoulders or half body. A tiny face in a wide shot gives a weak result.
- Sharp and evenly lit. Skip heavy filters, strong shadows and sunglasses.
- A stable composition with the subject in the centre leaves room for natural movement.
- Old photos: use the best scan you have. A cleaner, sharper scan helps the face hold up.
Audio tips
Clean audio gives the most precise mouth movement. Record somewhere quiet, close to the microphone, and avoid music underneath speech. For a song, a vocal with quiet backing works better than a loud full mix.
The video is exactly as long as the audio, so trim silence at the start and end before you upload. The track has to be 4 seconds to 5 minutes long, in MP3 or WAV.
If you want your own voice saying lines you never recorded, clone it once in the Audio studio from a short sample, then make the lines there and pick them in Lip Sync. Only clone your own voice or one you have written permission to use.
Writing the direction
The direction box is optional. Leave it empty and the voice alone drives the performance. Fill it in when you want a specific camera move, gesture or mood change.
Write a natural sentence in this order: camera, emotion, speaks or sings, actions, then anything happening around the subject. Say who is speaking ("the man speaks") and use "first ... then ..." for changes over time.
Example direction for a birthday message
Static medium close-up. The man speaks cheerfully, raises a glass towards the camera on "happy birthday", then laughs and waves at the end.
Ideas to try
Talking photos are useful far beyond novelty. A few that work well:
- A birthday or thank-you message in your own voice for someone far away.
- Your shop's founder photo introducing a new product.
- One portrait presenting every lesson of an online course.
- Your pet delivering a line for a social post.
- The same message in several languages from one photo.
- A character you designed telling its own story.
Consent and responsible use
Lip Sync makes the person in a photo say whatever the audio says, so use it only with permission. You must be the person in the photo or have their permission to make them speak, and you must have the right to use the voice and any music. On the website you confirm this the first time you generate with a new photo.
Never use it to deceive: no impersonation, no fake statements from real people, no harassment. Sexual content is never allowed, nor anything involving someone under 18, and photos with nudity or of minors are refused before anything is charged. When you share a realistic talking video of a real person, say that it is AI-generated. To report misuse, write to support@kreatorflow.ai.
Frequently asked questions
How do I make a photo talk for free?
Lip Sync is not free: it needs a paid plan or a credit pack, and costs 4 credits per second of audio at 720p. A 30-second message is 120 credits.
Can I make a photo talk with my own voice?
Yes. Upload a recording of yourself, or clone your voice in the Audio studio and make the line there. The recording is used as it is.
How long can the talking video be?
As long as the audio: 4 seconds to 5 minutes.
Can I make a photo of someone else talk?
Only with their permission, and only with a voice you have the right to use. Celebrities and public figures need their permission too, and nothing sexual or involving minors is allowed.
Can the photo sing?
Yes. Upload a vocal track and write "sings" in the direction. A clean vocal gives the best mouth movement.
Does it work with old photos and drawings?
Yes, if the face is clear. Old photos work best as a sharp scan; drawings, paintings, cartoon characters and animals work too.
Which resolution should I choose?
720p for tests and most social posts, 1080p for the final cut or a big screen. 1080p costs 6 credits per second instead of 4.
