SonicMaster: Towards Controllable All-in-One Music Restoration and Mastering
Paper • 2508.03448 • Published • 6
import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("AEmotionStudio/sonicmaster-models", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
SonicMaster — bundled mirror
Self-contained inference bundle for the MAESTRO desktop app.
Contents
sonicmaster/model.safetensors— diffusion checkpoint (source: amaai-lab/SonicMaster)sonicmaster/tangoflux_config.yaml— model architecture configt5/—google/flan-t5-largetext encoder + tokenizer (Apache-2.0)vae/— Oobleck VAE fromstabilityai/stable-audio-open-1.0
License
- SonicMaster: Apache-2.0
- T5 (flan-t5-large): Apache-2.0
- Stable Audio Open 1.0 VAE: Stability AI Community License (research-friendly)
Paper: Melechovsky, Mehrish, Herremans. "SonicMaster: Towards Controllable All-in-One Music Restoration and Mastering." ICML 2026 (arXiv:2508.03448).
- Downloads last month
- -
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support