ReferenceAsk
Back to Remy
Remy Reference/Models/Lipsync/Omni Human 1.5
F

Omni Human 1.5

Omni Human 1.5, a lipsync model from Fal.
lipsyncModelOverride: { model: "bytedance-avatar-omni-human-1.5" }
ByteDance·served viaFal·Released September 2025
Pricing
List price
$0.16/second
Per second
$0.16
Per minute
$9.60
Specifications
Context window
2,000 tokens
Aspect ratios
2
Released
September 2025
Model string
fal-ai/bytedance/omnihuman/v1.5
Parameters
Model options, passed in lipsyncModelOverride.config.
ParameterTypeDefaultRange / options
image
Portrait to animate into a talking video.
Image URL——
audio
Speech to lip sync to: up to 60 seconds at 720p, or 30 seconds at 1080p.
Audio URL——
prompt
Optional description of the expression and movement.
Prompt——
mask_image
Optional mask marking which person speaks: only the one in the white area moves their lips.
Image URL——
resolution
The resolution of the output video. 720p is faster and takes longer audio.
Select720p720p1080p
turbo_mode
Generate faster, with a slight loss of quality.
Booleanfalse
Use this model
await mindstudio.generateLipsync({
  lipsyncModelOverride: {
    model: "bytedance-avatar-omni-human-1.5",
    config: {
      image: "...",
      audio: "...",
      resolution: "720p",
    },
  },
});