Skip to main content
Your generations

AI lip sync and talking avatar generator

Make a photo talk: upload a portrait and a voice recording, and lip sync AI turns it into a talking video with natural head and face motion, for spokesperson clips, UGC-style ads, lessons and greetings.

  • One photo + a voice track
  • Lip-synced to your exact audio
  • Up to 5 minutes
  • From 8 credits a second

The avatar says exactly what is in the track. No recording? Make a voiceover in the Audio tab, download it and drop it here. Uploads are deleted after 7 days.

Priced per started second of the voice track. The video is as long as the track and is never watermarked. Only use photos and voices you have the right to use.

New accounts get 50 free credits

Made with MakeMidas

Real results from the models on this site, with the prompts that made them.

Kling AI Avatar 2.0

A generated late-night radio host lip-synced by Kling AI Avatar 2.0 to a Gemini voiceover ("It's almost midnight, and the city finally exhales.")

Popular models

Kling AI Avatar 2.0

Turns one photo and up to 5 minutes of voice into a 720p talking video.

About Kling AI

From 8 credits per second

Spokesperson and UGC-style ads

Give a product a presenter: a photo of your spokesperson or a generated character, your script as a voiceover, and a talking clip ready to cut into an ad.

Lip sync to the audio you bring

The avatar says exactly what is in your track, in whatever language it is spoken. Record yourself, or make a voiceover in the Audio tab and drop it in.

Short clips or long videos

InfiniteTalk is the cheapest for clips up to 15 seconds at 480p or 720p. Kling AI Avatar 2.0 makes 720p videos up to 5 minutes long.

Priced by the second

The voice track's length is measured on upload, and the button shows the price for the model and size you pick. Videos that fail are refunded.

Keep creating

One account, one credit balance for images, video and audio.

Choose your plan

A monthly plan or a one-time credit pack gives every model and clean output.

Black Friday: everyone gets 125 credits for every $1 on credit packs, up to 25% more credits. No subscription or code needed. Ends 2026-12-01 (UTC).
1,000 credits
$10one-time

Black Friday: 1,250 credits

  • Base credits: 1,000
  • Black Friday bonus: 250
  • Valid for 12 months
5,250 credits
$50one-time

Black Friday: 6,250 credits

  • Base credits: 5,250
  • Black Friday bonus: 1,000
  • Valid for 12 months
Recommended
11,000 credits
$100one-time

Black Friday: 12,500 credits

  • Base credits: 11,000
  • Black Friday bonus: 1,500
  • Valid for 12 months
28,750 credits
$250one-time

Black Friday: 31,250 credits

  • Base credits: 28,750
  • Black Friday bonus: 2,500
  • Valid for 12 months
62,500 credits
$500one-time

Subscribers get 62,500

  • Base credits: 62,500
  • Valid for 12 months

Any pack unlocks every model and several outputs at once, even without a subscription.

Questions

What do I need to make a talking avatar?

One clear photo with the face visible (JPG or PNG; WebP works with InfiniteTalk) and a voice track: MP3, WAV, M4A, AAC or OGG, up to 50 MB. InfiniteTalk takes up to 15 seconds of voice and Kling AI Avatar up to 5 minutes.

Can I start from a script instead of a recording?

Yes, in two steps: make a voiceover from your script in the Audio tab, download it, and upload it here. The avatar keeps that voice and its timing exactly.

How much does a talking avatar cost?

Kling AI Avatar 2.0: 8 credits per second at 720p. Clips are counted per started second of the voice track; a 10-second clip costs from 80 credits.

How long does it take?

Usually a few minutes for a short clip, longer for a video of several minutes. You can leave the page; the video appears in your history when it is ready, and uploads are deleted after 7 days.

Can I lip-sync an existing video?

Not yet. Lip sync here starts from a still photo, like a talking photo: the portrait is animated and its mouth follows your voice track. Re-syncing the mouth in footage you already have (video-to-video lip sync) is not offered.

Who owns the result?

The result is your own image or video, edited, so you can use it as you could use the original. Only upload files you have the right to use. Results stay in your history and only you can open them.

Make your first talking video