Apollo Attribution and Modified-Model Notice (CC BY-SA 4.0)

Project:
Apollo - Band-sequence Modeling for High-Quality Audio Restoration

Authors:
Kai Li, Yi Luo

Affiliations stated by upstream:
Tsinghua University; Tencent AI Lab

Copyright notice:
Copyright (c) Kai Li, Yi Luo. Apollo upstream is licensed by the authors under
Creative Commons Attribution-ShareAlike 4.0 International.

Disclaimer:
The Licensed Material and this adapted ONNX export are provided on an "as-is"
and "as-available" basis without warranties of any kind. See Section 5 of
model-LICENSE.CC-BY-SA-4.0.txt.

Upstream project:
https://github.com/JusperLee/Apollo

Upstream model checkpoint:
https://huggingface.co/JusperLee/Apollo/resolve/main/pytorch_model.bin

Upstream checkpoint SHA256:
99d9af7f1ff20e63c393035513a655392818d66b4d7fc23d658175c1f15e8d76

Upstream source snapshot tracked by delossify runtime metadata:
JusperLee/Apollo c46640ff65b2d42db6b890a99fd0bc95c5f50959

License:
Creative Commons Attribution-ShareAlike 4.0 International
https://creativecommons.org/licenses/by-sa/4.0/

Product model artifact:
The browser app expects the Apollo ONNX export served from /models/ or selected
locally by the user to match app/model-manifest.json.

- apollo_8s.onnx
  SHA256: 60c4c8c7f1f4e5012cae0ae81c074c0f84460d2e3d2e398dbfcf58efcef9802e
  Size: 78104875 bytes

Modified-model notice:
The ONNX file is a converted/exported derivative of the upstream Apollo PyTorch
checkpoint for fixed-window browser/WebGPU inference. It is not the original
upstream PyTorch checkpoint. Any public distribution of the ONNX file must keep
this attribution, the CC BY-SA 4.0 license reference, and a clear
notice that the model format and graph were changed for this browser runtime.

Upstream citation:
Li, Kai and Luo, Yi. Apollo: Band-sequence Modeling for High-Quality Music
Restoration in Compressed Audio. ICASSP 2025.