19 lip-sync engines · free plan · up to 10 min
Talking Photo AI — make any picture talk
Upload a portrait and a voice track, and the AI moves the lips, face and head in sync with the audio. 19 engines — Kling Avatar, OmniHuman 1.5, InfiniteTalk, HeyGen — with speech up to 10 minutes, on the free plan. Already have a video of someone speaking? AI lip sync matches it to new audio instead.
- Photo + your audio
- 19 engines
- Up to 10 min of speech
- Two-speaker engines
Updated by the Kenerate AI team
Made with Kenerate
On this page
Quick answer
What is a talking photo AI?
A talking photo AI animates a single portrait so its lips, face and head move in sync with audio, for presenters and social posts. In Kenerate AI's Avatar Talk you upload a photo and a voice track and pick one of 19 engines, with up to 10 minutes of speech and two-speaker options. It works on the free plan.
Key facts
- Engines
- 19
- Longest speech
- 10 min
- Two speakers
- Yes
- Free plan
- Yes
- Engine examples
- Kenerate Avatar, InfiniteTalk, Kling Avatar, OmniHuman 1.5
- Audio in
- MP3 or WAV, mic or library
Real results
Talking photo AI examples, with audio
Real Wan 3 clips with generated voices, from Seedream 5.0 Pro portraits. In Avatar Talk you bring your own audio for exact words.
How to
How to use Kenerate Talking Photo AI
- Time needed
- Steps
- 4 steps
Step 1
Open Avatar Talk
Sign in — it works on the free plan with sign-up credits.
Step 2
Upload a portrait
Facing the camera, face clearly visible, up to 20 MB. Cartoons and paintings work too.
Step 3
Add the voice
Upload MP3/WAV (up to 50 MB), record on the mic, or pick from your library. Need text read aloud? Make it in Voice Studio first.
Step 4
Pick an engine and generate
Kenerate Avatar for 1080p, InfiniteTalk or Kling Avatar for long speeches, a Multi engine for two speakers.
See it in the app
Open Avatar Talk, pick Talking photo, script to video or lip sync, and choose the engine.
Watch the tour · 2:23
Talking photos, script to video, engines, any language, presenters, lip sync & chat
Chapters
Engines
19 engines for talking photos
From Avatar Talk's engine list: how much speech each takes and what it's best at.
- Default
Kenerate
Kenerate Avatar
- Input
- Photo + up to 10 min audio
- Output
- 720p or 1080p
Open Kenerate AvatarBest for
Sharp 1080p talking photos
WaveSpeed
InfiniteTalk
- Input
- Photo + up to 10 min audio
- Output
- 480p or 720p
Open InfiniteTalkBest for
Long speeches
- 2 speakers
WaveSpeed
InfiniteTalk Multi
- Input
- Photo of two + two voices
- Output
- 480p or 720p
Open InfiniteTalk MultiBest for
Two people talking
Kuaishou
Kling Avatar 2 Pro
- Input
- Photo + up to 5 min audio
- Output
- Engine decides
Open Kling Avatar 2 ProBest for
Expressive presenters
Kuaishou
Kling Avatar 1 Pro
- Input
- Photo + up to 10 min audio
- Output
- Engine decides
Open Kling Avatar 1 ProBest for
Long Kling avatar videos
ByteDance
OmniHuman 1.5
- Input
- Photo + up to 60 s audio
- Output
- Engine decides
Open OmniHuman 1.5Best for
Portrait to full body, singing
All models (10)
Sync Labs
Lipsync-3
- Input
- Photo + up to 5 min audio
- Output
- Engine decides
Open Lipsync-3Best for
Precise lip sync
Meituan
LongCat Avatar
- Input
- Photo + up to 2 min audio
- Output
- 480p or 720p
Open LongCat AvatarBest for
Natural head motion
HeyGen
HeyGen Avatar
- Input
- Audio + one of 500 stock presenters
- Output
- 720p, 1080p or 4K
Open HeyGen AvatarBest for
No photo needed
The examples above were made with Wan 3 image to video from one prompt each. Avatar Talk uses your own audio, so the words are exactly yours.
Tips
Photos that talk naturally
What works best across the engines.
Face the camera, mouth closed, nothing covering the face.
Head-and-shoulders framing gives the most natural lip sync.
Use clean audio with no music under the voice.
Split very long scripts into chapters.
For singing, upload the vocal track without music underneath.
Only animate photos you have permission to use.
Example uses
How creators use talking photo AI
Illustrative examples of typical workflows, not customer reviews.
Museum educator
A museum educator can make a painted portrait speak a short introduction for a gallery guide.
Singer
A singer can upload a vocal track and make a portrait sing the line.
Course creator
A course creator can turn one portrait and a recorded script into a talking intro video.
FAQ
Talking Photo AI questions
8 questions
How do I make a picture talk with AI?
Open Avatar Talk, upload a portrait, add a voice track (upload, record or pick from your library) and generate. The engine moves the lips, face and head in sync with the audio.
Link to this answerCan I make a photo sing?
Yes — upload the vocal track as the audio. ByteDance shows its OmniHuman model animating singing from a single image; InfiniteTalk takes tracks up to 10 minutes.
Link to this answerCan I use my own voice?
Yes — record it on the mic or upload an MP3 or WAV. To have a script read aloud, make the voice in Voice Studio and use that file.
Link to this answerWhat kind of photo works best?
A face looking at the camera, mouth closed, nothing covering it, with head and shoulders in frame. Paintings, cartoons and 3D characters work too.
Link to this answerHow long can a talking photo video be?
Up to 10 minutes of speech on Kenerate Avatar, InfiniteTalk and Kling Avatar 1 Pro; 5 minutes on Kling Avatar 2, Lipsync-3 and Fabric; 60 seconds on OmniHuman 1.5.
Link to this answerIs talking photo AI free?
Avatar Talk works on the free plan — use your free sign-up credits. Longer videos use one-time credit packs that never expire.
Link to this answerCan I make a celebrity's photo talk?
No. Only animate photos of yourself, characters you made, or people who agreed to it.
Link to this answerWhat's the difference between a talking photo and lip sync?
A talking photo starts from one still portrait and generates the whole performance — lips, face, blinks and head movement — from your audio. Lip sync starts from an existing video and only redraws the mouth to match a new track, keeping the original performance. Have a video already? Use AI lip sync. Have only a picture? Use talking photo AI.
Link to this answerIn depth
Talking Photo AI, explained
On this page
How talking photos work
The engine turns the audio into mouth shapes and facial motion, then animates your portrait frame by frame to match — adding blinks and head movement so it doesn't look frozen. ByteDance's OmniHuman animates a single image from audio, video or both, from portrait to full body.
Talking photo vs AI presenter
A talking photo animates your own picture. HeyGen-style presenters use a stock avatar instead — Kenerate's HeyGen Avatar engine offers 500 of them, so you only need the audio.
Photo checklist
D-ID recommends a camera-facing medium shot with a neutral expression and mouth closed, the face at least 200 × 200 px, and nothing covering it. The same rules help every engine.
Explore
Every Kenerate video tool
Generators, tools, models and guides — all in one studio.
By the Kenerate AI teamLast updated and reviewed
Sources: Kenerate video studio · HeyGen — Avatar IV guide · Kling — AI Avatar 2.0 guide · OmniHuman-1 project page · D-ID — FAQs
