New Kling 3.0 generates cinematic video with synchronized audio.Try free

Make a portrait speak

Upload a photo or cast a saved identity, then supply audio or write a script. The video is as long as the speech, and billed per second of it.

Who is speaking

Newidentity

Upload a portrait, or create an identity to reuse one across studios.

…or paste an image URL

What they say

The output matches this audio's duration.

Consent

Animating a real person's likeness requires their permission. Choose which applies — neither is selected by default.

This is meI affirm that the portrait I am uploading is of myself, that I consent to Draftless animating my likeness so that it appears to speak, and that I understand the generated video may be indistinguishable from real footage of me.This is someone else, with their permissionI affirm that the portrait I am uploading is not of myself, that the person depicted has given me explicit, documented permission to animate their likeness so that it appears to speak, and that I am not animating any person's likeness without their permission.

Consent is required before generating.

Billed per second of audio (~34 cr/s). The exact reserve is confirmed before anything runs.

ExampleSynthetic presenter — not a real person

Source portrait of the synthetic example presenter — Portrait

Talking video

A photo and a voice become a talking video.

Cast someone, then give them something to say

Upload a portrait or pick a saved identity on the left. Supply an audio track, or write a script and we speak it in their voice.