Each clip is one photo plus one voice recording. Turn the sound on to hear the lip sync, or hit Try this example to load the same photo and voice.
An AI talking photo is a short video made from one still picture and one voice clip. The AI reads the sound of each word and moves the lips, jaw and cheeks to match, then adds the small things a real speaker does: blinks, eyebrow raises, a nod, a smile between sentences.
Talking photo AI keeps what is in your picture. The face, hair, clothes and background stay the same; only the person comes to life. The voice you hear is your clip, unchanged, so the result sounds exactly like the recording you uploaded.
Upload an MP3, WAV or M4A of the words you want said: your own recording, a voice note or an MP3 from the AI Voice Generator. Trim it to 4 to 60 seconds.
Pick a photo with one clear face turned toward the camera. It can be you, a friend who said yes, a painting, a cartoon character or your pet.
Press Create. In about two minutes the face speaks your clip in sync, with your audio as the soundtrack. Watch it in the browser and download the MP4.
The mouth follows every word of your clip. Nothing is re-voiced, so accents, laughs and pauses come through as recorded.
Say a full intro, a greeting or a short story in one take, not just a single line.
Blinks, eyebrows, small head moves and smiles follow the speech, so it looks like talking, not a moving mouth on a still face.
A portrait photo gives a 9:16 video for TikTok and Reels, a wide photo gives 16:9 for YouTube, and a square photo stays square. Videos are 720p.
A front or slightly turned face gives the cleanest lip sync. Side profiles and faces looking away are hard to animate.
Keep hands, microphones, masks and hair away from the lips, so the AI can see what it needs to move.
A clear photo at least 300 pixels on each side works best. Blurry or very dark faces lose detail when they move.
A portrait or upper-body shot keeps the face large in the frame. Tiny faces in a wide group photo come out soft.
A clear voice with little background noise gives the sharpest mouth shapes. Loud music under the voice can blur them.
The whole clip drives one face. If two people talk, the photo will say both parts.
Lip sync follows the sound, not the text, so English, Spanish, Japanese or any other language works.
Write your script in the AI Voice Generator, pick a voice or clone your own, download the MP3 and upload it here. Singing works too, see the AI Music Video Generator.
Three ways to bring a photo to life on ImagesArt. Pick by what you want to hear.
Turn a headshot into a presenter for a product demo, a course lesson or a channel intro.
Let a grandparent's photo say happy birthday, or record a message from a loved one in their own voice.
Make a cartoon, a game character or a brand mascot talk for social posts and ads.
Give your cat or dog a voice and post the clip on TikTok, Reels or Shorts.
With several people in the photo, one face is animated. Crop to the person you want first.
The background and camera stay still. To move the scene, use Image to Video.
There is no script box on this page. Record yourself, or turn text into speech with the AI Voice Generator first.
Upload your own photo, or someone who said yes. Do not make anyone appear to say things they never said.
A talking photo costs 14 credits per second of voice: 84 credits for a 6-second line, 210 for 15 seconds, 840 for the 60-second maximum. It runs at 720p, which needs a Starter plan or any credit pack. Failed runs are not charged.
Cancele quando quiser · Mantenha o acesso até o fim do período pago
Add a voice clip and a photo, and get a lip-synced talking video in about two minutes.