Official BytePlus partnerBytePlus, a ByteDance company
Official BytePlus partner · Singing photos
Make a photo sing: one picture, one song
To make a photo sing, upload the picture and the vocal track to KreatorFlow Lip Sync and press Generate. OmniHuman 2.0 performs the song, with the lips on the lyrics and the face following the melody, in one take of up to 5 minutes.
- Ballads to fast rap
- Songs up to 5 minutes
- People, characters and pets
- 720p or 1080p
- 4 credits per second
Last updated
On this page
An AI singing photo, performed rather than mouthed
Lip Sync on KreatorFlow runs ByteDance's OmniHuman 2.0, an audio-driven model: the sound leads the whole performance. On a song that means more than moving lips. Long notes hold the mouth open, breaths show between lines, and the head and eyebrows rise with the chorus. KreatorFlow is an official partner of BytePlus, a ByteDance company and runs the model through BytePlus.
The video is exactly as long as your track, so a whole song can be one continuous take with the same face from the first line to the last.
How to make a photo sing in three steps
- 1
Choose the singer
Upload a photo of yourself, of someone who agreed, or of a character or pet. A face turned towards the camera with the mouth fully visible gives the cleanest singing.
- 2
Add the song
Upload the vocal track as an MP3 or WAV, 4 seconds to 5 minutes long. A verse or chorus works for a short clip; most full songs fit in one take.
- 3
Generate the performance
Add an optional direction (eyes closing on long notes, a slow push-in), choose 720p or 1080p and press Generate. The price shows first: 4 credits per second at 720p.
Tips for a believable singing photo
- Clean vocals first. A vocal-only stem is ideal. If you only have the full mix, choose a version where the music is quieter.
- Pick a photo with energy. A relaxed, slightly open mouth and a lifted chin read as singing better than a tight smile.
- Trim the intro. Long instrumental intros become seconds of a silent face. Cut the track to start near the first line, and you pay for fewer seconds.
- Direct the emotion. Use the direction for what the audio cannot show, for example “She closes her eyes on the long notes and sways slightly to the beat.”
- Test the chorus. Make a short clip of the hardest part at 720p before you render the full song at 1080p.
What people make with singing photos
Music video visuals
A singer, real or virtual, performing your track for a lyric video or visualiser.
Virtual artist
A character that sings every release in the same voice and the same face.
Happy birthday song
A photo of you singing happy birthday to someone far away, in your own recorded voice.
Singing pet
Your cat or dog performing a song you have the rights to, for a funny reel.
Cover preview
A quick performance clip of your own cover recording, before a full shoot.
Rap verse
A character delivering a fast verse, with the mouth keeping up.
More lip sync examples
Real results, each from one picture and one voice track or song.
Two people, one speaker
“Okay, so here's the thing nobody tells you about starting a podcast. The first ten episodes? Nobody listens…”
14 seconds at 720p · English · The Jonah voice
A talking dog
“Okay, I need everyone to stay calm. I did not eat the sandwich. The sandwich simply left.”
6 seconds at 720p · English · The Jonah voice
A painting that speaks French
“Trois cents ans immobile dans ce musée, à sourire aux touristes. Enfin, quelqu'un me laisse parler. Merci, mes amis.”
7 seconds at 720p · French · A French Text to Speech voice
An anime character in Japanese
“こんばんは！雪がきれいですね。今日はずっと、あなたに会いたかったんです。”
6 seconds at 720p · Japanese · A Japanese Text to Speech voice
Real results from KreatorFlow Lip Sync, each made from one AI-generated picture (no real people) and audio from our Audio studio. Press play to hear them. AI-generated video.
What a singing video costs
Credits per started second of audio, from the same code the studio charges with.
| 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 typical three-minute song is 720 credits at 720p. Lip Sync needs a paid plan or a credit pack, and credits for a video that fails come back automatically.
Consent and responsible use
Lip Sync makes the person in a photo say whatever the audio says, so it comes with clear rules. On the website you confirm, the first time you generate with a new photo, that you are that person or have their permission. Photos are screened before anything is charged, and photos with nudity or of minors are refused.
- Use a photo of yourself, or of someone who has given you permission to make them speak. A character you created, an animal or a drawing is fine.
- Use a voice you have the right to use: your own, one of our voices, a voice you cloned with consent, or a recording you have permission for. The same goes for any music or song.
- No sexual content, nothing involving anyone under 18, and no impersonation: never make a real person appear to say something they did not say in order to mislead, harass or defraud.
- No celebrities, politicians or other public figures unless they have given you permission and you hold the rights for that use.
- When you post a talking video of a real person, say that it is AI-generated. Many platforms require that label.
Read the full rules in the Acceptable Use Policy on real people's faces and the Terms.
Singing photo questions
How do I make a photo sing?
Upload the photo to KreatorFlow Lip Sync, add the song as an MP3 or WAV and press Generate. OmniHuman 2.0 follows the vocal: the mouth matches the lyrics, and the face, breathing and head move with the melody and emotion.
Does it work with rap and fast songs?
Yes. The model follows fast rap verses as well as slow ballads. The faster the lyrics, the more a clean vocal matters, so use a version where the voice is clear.
Do I need the vocals separated from the music?
No, but it helps. A vocal-only track, or a mix where the voice is much louder than the music, gives the most accurate mouth movement. With a loud full mix, drums and other instruments can blur the timing.
How long can the song be?
4 seconds to 5 minutes per video. Most pop songs fit in one take. For a longer track, split it into parts and join the videos in an editor.
Can I use any song?
Only songs you have the rights to use: your own recording, an original song you made, or a track you licensed for this use. Posting a famous song can get the video muted or removed on social platforms, and can infringe the rights holder's copyright.
How much does a singing photo cost?
4 credits per second at 720p and 6 at 1080p. A one-minute song costs 240 credits at 720p, and the longest take (5 minutes) costs 1,200. You see the price before you generate.
Is there a free version?
No. Singing photos use Lip Sync, which needs a paid account: any plan, from $15 for the first month on Creator, or a credit pack from $8. Credits for a failed video come back automatically.
Can a cartoon, anime character or pet sing?
Yes. Any picture with a clear face and mouth can sing: drawn and 3D characters, anime characters, paintings and animals.
Make your photo sing
Add a photo and a song you have the rights to, and see the price before you generate. Available on any paid plan or credit pack.