AI talking avatar — your face, a character or a stock presenter

Upload a clear, front-facing photo of yourself in Avatar Talk, add your voice by recording it or uploading a file, and generate with Kenerate Avatar. Keep that photo as your master avatar image and pair it with new audio for each new video.

Yes. Choose HeyGen Presenters in the engine picker and pick one of 500 stock presenters in settings such as an office, a sofa, a gym or a doctor's room. You only add the audio, and the video comes out in 16:9 or 9:16 at up to 4K.

Yes. Upload the character image as the photo; cartoons, 3D renders and paintings all work as long as the face is clear and looks toward the camera. Make the character in an image generator first if you don't have one.

On 14 of the 19 talking-photo engines, yes: an optional direction box lets you describe gestures, mood and framing, with presets such as "warm and friendly, gentle nods". SkyReels V3 also takes emotion and camera moves. Engines without a prompt move the avatar from the audio alone.

Yes. Put both characters in one image and use InfiniteTalk Duo, InfiniteTalk Fast Duo or LongCat 1.5 Duo; each takes one audio track per person, played together or in turn.

Avatar Talk works on the free plan with the sign-up credits, so you can make your first talking avatar without buying anything. Longer or higher-resolution videos use more, and one-time credit packs top it up.

Only with their permission. Use your own face, characters you created, stock presenters, or people who agreed to be animated.