NovaSR2026voice
NovaSR Speech Upscaler - NovaSR
Model card for the NovaSR restoration preset in Neural Analog (NovaSR, 2026)
Available in Studio. Uses standard processing minutes.
NovaSR Description
Upscaling model trained on English speech. Best for restoring podcasts, voice-overs, or isolated vocal tracks. Output sample rate: 48kHz.
NovaSR Parameters
| Key | Label | Options | Description |
|---|---|---|---|
| stereo_mode | Channels | Mono (single_pass), Stereo (Mid/Sides) (mid_sides) | No additional help text. |