LavaSR v22026voice

LavaSR Speech Upscaler - LavaSR v2

Model card for the LavaSR v2 restoration preset in Neural Analog (LavaSR v2, 2026)

Available in Studio. Uses standard processing minutes.

LavaSR v2 Description

Fast and high-quality speech upscaling. Evolution of NovaSR that uses Vocos architecture for efficiency. Output sample rate: 48kHz.

LavaSR v2 Parameters

KeyLabelOptionsDescription
stereo_modeChannelsStereo (single_pass), Stereo (Mid/Sides) (mid_sides)No additional help text.