htdemucs 4-source weights, ggml format

HT-Demucs (Demucs v4) weights converted to the ggml binary format used by demucs.cpp, for running music source separation in WebAssembly.

Re-hosted so Audio Magic has a pinned copy it controls. Nothing has been modified โ€” byte-identical to the upstream conversion.

Provenance

Original model Demucs v4 / HT-Demucs โ€” Meta Platforms, MIT
Inference engine demucs.cpp โ€” Sevag Hanssian, MIT
ggml conversion Retrobear/demucs.cpp, revision 8f58ac0491bbea657275bcd3e38af1ab3a27bfc9
File ggml-model-htdemucs-4s-f16.bin โ†’ ggml-htdemucs-4s-f16.bin
Size 83,994,361 bytes
SHA-256 72b17c42d308982ddb5069bc3bf48b81a5aac4cb6516e4366c0fa7cef6df0064

Stems, in output order: drums, bass, other, vocals. Weights are f16.

Verification

Loaded into a demucs.cpp WebAssembly build and run against a real 44.1 kHz stereo recording: the model initialises in 0.2 s and the four stems reconstruct the input at 28.5 dB SNR.

Licence

Demucs and its weights are MIT (Meta Platforms); demucs.cpp is MIT (Sevag Hanssian). Both notices are included as LICENSE-demucs and LICENSE-demucs.cpp.

Citation

@inproceedings{rouard2023hybrid,
  title     = {Hybrid Transformers for Music Source Separation},
  author    = {Rouard, Simon and Massa, Francisco and D{\'e}fossez, Alexandre},
  booktitle = {ICASSP 23},
  year      = {2023}
}
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support