UniverSR Upscaler - UniverSR
Model card for the UniverSR restoration preset in Neural Analog (UniverSR, 2026)
Available in Studio as a 30s preview on Free and Plus. Full-length processing is available on Pro and Max. Consumes 6x the processing minutes. Maximum input length is 30 minutes.
UniverSR Description
Upscale music, voice, and sound effects. UniverSR is a 2026 model developed by the University of Seoul which performs audio super-resolution directly in the complex STFT domain using flow matching. Very similar to AudioSR, but with more coherent high ends. Recommended for: Low quality recordings, missing high frequencies, hissing noise around 10Khz (hihats, voice, suno hiss). Low thresholds (eg: 4Khz) will change audio the most, while high threshold will preserve what's already here. Not recommended for: muddy bass or muddy low end. Use Neural Remix instead.
UniverSR Architecture
UniverSR Parameters
| Key | Label | Options | Description |
|---|---|---|---|
| frequency_cutoff | Frequency Cutoff | 12Khz (SR: 24Khz) (12000), 8Khz (SR: 16Khz) (8000), 6Khz (SR: 12Khz) (6000), 4Khz (SR: 8Khz) (4000) | No additional help text. |
| stereo_mode | Channels | Stereo (Mid/Sides) (mid_sides), Stereo (L/R) (left_right), Mono (single_pass) | No additional help text. |
| reconstruction_method | Reconstruction | Seamless Reconstruction (original_signal), Original Reconstruction (original) | No additional help text. |