Make a photo talk

Upload a portrait, add a voice you record, upload or type, and get a lip synced talking video. 2 credits a second, no watermark on any plan, and you own the result.

Make a talking photo

Real talking photo examples

Output from InfiniteTalk Fast, the lip sync model behind Percify's talking photos: one still image and one audio track in, each clip under 10 seconds. Press play for sound.

Check your photo first, free

The checker runs in your browser and tells you whether the face is big enough, facing the camera, lit and sharp. Nothing is uploaded until you choose to make the video.

Will this photo lip sync well?

Choose a photo and the checker finds the face, then looks at its size, angle, light and focus. It runs in your browser; the photo is not uploaded.

How it works

  1. Step 1

    Choose a photo

    A sharp, front facing head and shoulders photo of one person. A render or an illustrated face with clear eyes and mouth works too.

  2. Step 2

    Add the voice

    Record it, upload an audio file, or pick one of your saved voices or a public voice and type what it should say.

  3. Step 3

    Generate and download

    The mouth, jaw and head move with the audio. Download the MP4 with no watermark.

What it costs

Talking photos are priced per second of video. Runs that fail are refunded.

ModelCredits a second15 s clip60 s clip
InfiniteTalk Fast230120
InfiniteTalk, 480p460240
InfiniteTalk, 720p690360

Credit prices by plan are on the pricing page. Checked 17 September 2026.

Talking photo apps compared

The question is rarely whether a tool can make a photo talk. It is what the free plan leaves on the video and what you are allowed to do with it.

ToolFree planWatermark removed from
PercifyNo watermark, commercial use allowedNever added, on any plan
HeyGenWatermarked, personal and evaluation use onlyCreator, $29 a month
D-ID14 day trial with a full screen watermark, personal useOnly Enterprise removes it fully
VidnozCommercial licence, but with its watermarkStarter, $26.99 a month
HedraMay be watermarked, non commercialBasic, $15 a month

Read from each vendor's own pricing page on 17 September 2026; prices change, so check theirs before relying on a figure. The full table with sources is on best AI avatar generator. Percify is our product.

Watermarks, rights and consent

  • No watermark on any plan, including the free one.
  • You own the content you generate, under Percify's terms. It is not exclusive: similar inputs can give similar output for someone else.
  • Use your own photo and voice, or ones you have permission to use.

Details, and how to check any other tool for hidden watermarks: AI avatar watermarks and commercial use.

Make it sound like you

For a video in your own voice, record a clean sample once and reuse it for every script. Check the recording with the free voice sample checker, then set up Clone Yourself with your photo and voice.

Make it move instead of speak

Lip sync animates the mouth to a voice. If you want the picture itself to move, with the subject and the camera in motion and no speech, that is image to video: 23 models from 3 credits a second, with the price of a five second clip on each one.

Neither of those is lip reading, which is working out what a silent speaker said. Percify does not do that, and nor does anything else reliably yet; the apps that try are compared in the lip reading app comparison.

Animate a photoAI image generatorAI headshot generatorAI avatar generatorAudio to videoAI video generatorText to videoTalking photo checkerVoice sample checkerScript length calculatorLip sync apps comparedStep by step guide in the docs

Make your photo talk

One photo and one voice in, a lip synced MP4 out. 2 credits a second, no watermark.

Make a talking photo

Frequently asked questions

Everything you need to know before you start.

Upload a clear, front facing photo, add the voice (record it, upload an audio file, or pick a voice and type what it should say), and generate. Percify animates the mouth, jaw and head to the audio and gives you an MP4 to download.

It depends on what you need: realistic lip sync on your own photo, your own cloned voice, no watermark, commercial use, or an API. Percify covers those on every plan and charges per second of video, but it is our product, so compare it on the same points with the other tools you are considering.

Yes. Give it a still photo and an audio track and it returns an MP4 of that person speaking, with the mouth, jaw and head moving to the voice. A 10 second clip is 20 credits on the fast model, roughly 35 to 50 cents.

A talking photo is a talking avatar built from your own photo instead of a preset character. Both produce a video of a face speaking your script; the difference is where the face comes from. Upload a photo here, or pick a ready made presenter if you would rather not use your own.

InfiniteTalk Fast costs 2 credits a second: a 15 second clip is 30 credits and a 60 second clip is 120. Credits work out at about 2 cents each depending on the plan, so a minute of talking video costs roughly $2 to $3. The higher quality InfiniteTalk model costs 4 credits a second at 480p or 6 at 720p. Runs that fail are refunded.

No. Talking photos come out clean on every plan, including the free one, and there is no paid upgrade to remove a mark because there is no mark. The one Percify tool that does mark its output is Replicate a short, whose replicas carry a small corner mark as an AI-content disclosure; that is a different generator from this one.

Yes. Percify’s terms say you own the AI generated content you create, subject to the terms. Output is not exclusive, and you need the rights to the photo and the voice you use.

Yes. Record or upload your voice for a clip, or save it and type new scripts for it. The free voice sample checker tells you if a recording is clean enough first.

A sharp, well lit, front facing head and shoulders photo of one person. The free photo checker on this page tests size, angle, tilt, light and focus before you spend anything.

Your own, or a photo of someone who agreed to it. Percify’s terms require that you have the rights to what you upload.