Nodes/ComfyUI/LTXV Reference Audio (ID-LoRA)
ComfyUI Node Runs on cloud

LTXV Reference Audio (ID-LoRA)

Set reference audio for ID-LoRA speaker identity transfer. Encodes a reference audio clip into the conditioning and optionally patches the model with identity guidance (extra forward pass without reference, amplifying the speaker identity effect).

By Comfy-Org·Created 4 years ago·Updated 20 days ago· 121,575
LTXV Reference Audio (ID-LoRA)
  • model
  • positive
  • negative
  • reference_audio
  • audio_vae
  • MODEL
  • positive
  • negative
identity_guidance_scale3.00
start_percent0.000
end_percent1.000
Categorymodel/conditioning/ltxv

Inputs (8)

NameTypeDefaultDescription
modelMODEL
positiveCONDITIONING
negativeCONDITIONING
reference_audioAUDIOReference audio clip whose speaker identity to transfer. ~5 seconds recommended (training duration). Shorter or longer clips may degrade voice identity transfer.
audio_vaeVAELTXV Audio VAE for encoding.
identity_guidance_scaleFLOAT3.000–100Strength of identity guidance. Runs an extra forward pass without reference each step to amplify speaker identity. Set to 0 to disable (no extra pass).
start_percentFLOAT0.0000–1Start of the sigma range where identity guidance is active.
end_percentFLOAT1.0000–1End of the sigma range where identity guidance is active.

Outputs (3)

NameTypeDescription
MODELMODEL
positiveCONDITIONING
negativeCONDITIONING