Logo
Image
Video
Voice
Motion Control
Kling EffectsNew
CloningNew
WorkflowNew
Character
Dubbing
Apps
More
Zyka Foundry
Refer & Earn
Pricing

Turn your idea into video

Ready to create some AI magic? Describe your scene, pick your settings, and watch as AI brings your vision to life in stunning video.

Write a detailed prompt describing your video scene
Optionally upload reference images or videos
Hit generate and create cinematic videos in minutes
Start creating on the left
fal.ai

OmniHuman — bring human portraits to life with audio

OmniHuman v1.5 from ByteDance, available on Zyka, generates lifelike human videos from a single portrait image and an audio file. Perfect for talking-head content, voiceover videos, and avatar animation.

  • Image + audio to video
  • Lifelike facial animation
  • Lip-sync to provided audio
  • Single-image input

How OmniHuman v1.5 Works

01

Upload a portrait

Provide a clear image of a human figure — OmniHuman uses it as the visual identity for the generated video.

02

Provide an audio file

Upload the voice audio you want the subject to speak. OmniHuman lip-syncs the portrait to the audio.

03

Generate the talking video

OmniHuman renders the portrait speaking the audio with lifelike facial animation and natural lip-sync.

About OmniHuman v1.5

OmniHuman v1.5 is ByteDance's specialized image-and-audio-to-video model on Zyka. It animates a single human portrait to speak provided audio with realistic lip-sync and facial motion.

Compared to general-purpose video models, OmniHuman is purpose-built for talking-head content — making it more accurate and consistent for that specific use case than text-to-video models attempting the same task.

Use OmniHuman on Zyka for voiceover videos, AI avatar content, narrated explainers, and any scenario where you need a portrait to deliver spoken audio convincingly.

Frequently Asked Questions

What is OmniHuman?

ByteDance's specialized model for generating talking-head video from a single portrait image and an audio file.

What kind of portrait works best?

Clear, well-lit, front-facing portraits with the subject's mouth visible. Avoid heavy occlusion or extreme angles.

Does it lip-sync to my audio?

Yes. OmniHuman generates lip motion synchronized to the provided audio track.

How is OmniHuman different from Aurora or Infinite Talk?

All three target talking-head generation. OmniHuman, Aurora, and Infinite Talk each have different model lineages and aesthetics — try them on the same input to see which suits your project.