Runtime models, model card and licences

Evaluation/LICENSE-MDX23C-Code.txt

@@ -0,0 +1,31 @@
+MIT License
+
+mdxnet-infer's model.py is a verbatim-ported, inference-only subset of
+ZFTurbo's Music-Source-Separation-Training (https://github.com/ZFTurbo/Music-Source-Separation-Training),
+whose MDX23C TFC-TDF architecture originates from KUIELab's sdx23
+(https://github.com/kuielab/sdx23/). Both are MIT licensed; both are
+credited below alongside OpenMIRLab's packaging of this inference-only port.
+No model weights are bundled with this software — see README.md for weights
+provenance and licensing, which is separate from the code license below.
+
+Copyright (c) 2023 KUIELab
+Copyright (c) 2023 Roman Solovyev (ZFTurbo)
+Copyright (c) 2024 OpenMIRLab
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

Evaluation/mdx23c-drums.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:99d68124a0eabb3291bf155334fb55ee0af1bc9957d54bd2167a7a3458aa7070
+size 437702629

Evaluation/mdx23c-native.json

@@ -0,0 +1,42 @@
+{
+  "name": "MDX23C six-part experimental drum refinement",
+  "file": "mdx23c-drums.onnx",
+  "sha256": "99d68124a0eabb3291bf155334fb55ee0af1bc9957d54bd2167a7a3458aa7070",
+  "bytes": 437702629,
+  "input_shape": [
+    1,
+    4,
+    1024,
+    256
+  ],
+  "output_shape": [
+    1,
+    6,
+    4,
+    1024,
+    256
+  ],
+  "roles": [
+    "Kick",
+    "Snare",
+    "Toms",
+    "Hi-Hat",
+    "Ride",
+    "Crash"
+  ],
+  "source_checkpoint_sha256": "d2a4aa53eb584d21eead358a4e66d1882ad182911be018f052b5da73be9096d0",
+  "source_config_sha256": "17d1649a227f841165bdb4c11a42082898192a1ea3ceab7e7e0b9293d6589dd6",
+  "code_revision": "cb4779eedeca4f55ba498df8e234e2b5f4b90aaf",
+  "conversion": "Scripts/evaluation/export_mdx23c.py; opset 17, float32, fixed 256-frame input",
+  "status": "Local evaluation only; original checkpoint redistribution terms unresolved.",
+  "source_repository": "https://github.com/openmirlab/mdxnet-infer",
+  "export_environment": {
+    "torch": "2.8.0",
+    "onnx": "1.23.2",
+    "onnxruntime": "1.23.2",
+    "numpy": "2.2.6",
+    "PyYAML": "6.0.3"
+  },
+  "source_checkpoint": "Public model repository, release weights-drumsep-v1: aufr33-jarredou_DrumSep_model_mdx23c_ep_141_sdr_10.8059.ckpt (authors aufr33 and jarredou)",
+  "source_config": "Same release: aufr33-jarredou_DrumSep_model_mdx23c_ep_141_sdr_10.8059.yaml"
+}

Fast/Models/htdemucs_6s.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d298f7f746bf53c21baad44fb08e88807ef47feb551dd22f1601a546c85b8e02
+size 109726583

Fast/Models/htdemucs_6s_config.json

@@ -0,0 +1,94 @@
+{
+  "model_name": "htdemucs_6s",
+  "model_class": "BagOfModelsMLX",
+  "sub_model_class": "HTDemucsMLX",
+  "num_models": 1,
+  "weights": [
+    [
+      1.0,
+      1.0,
+      1.0,
+      1.0,
+      1.0,
+      1.0
+    ]
+  ],
+  "args": [],
+  "kwargs": {
+    "sources": [
+      "drums",
+      "bass",
+      "other",
+      "vocals",
+      "guitar",
+      "piano"
+    ],
+    "audio_channels": 2,
+    "samplerate": 44100,
+    "segment": "39/5",
+    "channels": 48,
+    "channels_time": null,
+    "growth": 2,
+    "nfft": 4096,
+    "wiener_iters": 0,
+    "end_iters": 0,
+    "wiener_residual": false,
+    "cac": true,
+    "depth": 4,
+    "rewrite": true,
+    "multi_freqs": [],
+    "multi_freqs_depth": 3,
+    "freq_emb": 0.2,
+    "emb_scale": 10,
+    "emb_smooth": true,
+    "kernel_size": 8,
+    "stride": 4,
+    "time_stride": 2,
+    "context": 1,
+    "context_enc": 0,
+    "norm_starts": 4,
+    "norm_groups": 4,
+    "dconv_mode": 3,
+    "dconv_depth": 2,
+    "dconv_comp": 8,
+    "dconv_init": 0.001,
+    "bottom_channels": 0,
+    "t_layers": 5,
+    "t_hidden_scale": 4.0,
+    "t_heads": 8,
+    "t_dropout": 0.02,
+    "t_layer_scale": true,
+    "t_gelu": true,
+    "t_emb": "sin",
+    "t_max_positions": 10000,
+    "t_max_period": 10000.0,
+    "t_weight_pos_embed": 1.0,
+    "t_cape_mean_normalize": true,
+    "t_cape_augment": true,
+    "t_cape_glob_loc_scale": [
+      5000.0,
+      1.0,
+      1.4
+    ],
+    "t_sin_random_shift": 0,
+    "t_norm_in": true,
+    "t_norm_in_group": false,
+    "t_group_norm": false,
+    "t_norm_first": true,
+    "t_norm_out": true,
+    "t_weight_decay": 0.0,
+    "t_lr": null,
+    "t_sparse_self_attn": false,
+    "t_sparse_cross_attn": false,
+    "t_mask_type": "diag",
+    "t_mask_random_seed": 42,
+    "t_sparse_attn_window": 400,
+    "t_global_window": 100,
+    "t_sparsity": 0.95,
+    "t_auto_sparsity": false,
+    "t_cross_first": false,
+    "rescale": 0.1
+  },
+  "mlx_version": "0.30.3",
+  "tensor_count": 565
+}

Guitar/LICENSE-MSS-Code.txt

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2024 Roman Solovyev (ZFTurbo)
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

Guitar/guitar-evaluation.json

@@ -0,0 +1,53 @@
+{
+  "scope": "Intake receipt for the shipped guitar model (BS-RoFormer SW, fixed). Licence status: unresolved; no licence is published for the weights.",
+  "files": [
+    {
+      "repository": "enerjazzer/BS-ROFO-SW-Fixed",
+      "revision": "a443a2985534b3bc815ef54a5d446c6a0390f974",
+      "file": "BS-Rofo-SW-Fixed.ckpt",
+      "bytes": 699412152,
+      "sha256": "24e7d35ee9c64415673d3fd33e06a67cac2c103c5df6267ba1576459c775916e",
+      "publisher_sha256": "24e7d35ee9c64415673d3fd33e06a67cac2c103c5df6267ba1576459c775916e"
+    },
+    {
+      "repository": "enerjazzer/BS-ROFO-SW-Fixed",
+      "revision": "a443a2985534b3bc815ef54a5d446c6a0390f974",
+      "file": "BS-Rofo-SW-Fixed.yaml",
+      "bytes": 4613,
+      "sha256": "f9fada9f94e5ba2d2e4600196299459294bc5f532b314c209cc156ac63e4329b",
+      "publisher_sha256": null
+    },
+    {
+      "repository": "enerjazzer/BS-ROFO-SW-Fixed",
+      "revision": "a443a2985534b3bc815ef54a5d446c6a0390f974",
+      "file": "README.md",
+      "bytes": 95,
+      "sha256": "5c717c20c694c9489fde0909482df6137639d890c8c54bf318ef83d09b0ce0a6",
+      "publisher_sha256": null
+    }
+  ],
+  "upstream_architecture_revision": "6fdad13686bb7e84d029ec66f9dbe76e65c93ebf",
+  "architecture_sha256": {
+    "Development/RoFormer/source/models/bs_roformer/__init__.py": "fc236d3158ba954e77ec7f7d4cae8eec924bebc3150db4885ab946d693f1e862",
+    "Development/RoFormer/source/models/bs_roformer/attend.py": "c7abbc40a3fd20ff7f6001fa9f8ee9ad5df6452272712a5e719b8f8468bf2223",
+    "Development/RoFormer/source/models/bs_roformer/bs_roformer.py": "a6f670325cdb8a7a212914f62602119b23eb888df9bcfe4a85f9be74728b3ef4",
+    "Development/RoFormer/source/models/bs_roformer/mel_band_roformer.py": "a28fce62020faad6461176660e012cf5799ae8a26c76fe4b2bc636f5b44b9fc5"
+  },
+  "adapter_checks": [
+    {
+      "model": "SW",
+      "device": "cpu",
+      "max_abs": 0.0,
+      "relative_rms": 0.0,
+      "rotary_policy": "checkpoint"
+    },
+    {
+      "model": "SW",
+      "device": "mps",
+      "max_abs": 1.2069940567016602e-06,
+      "relative_rms": 4.315405476518208e-06,
+      "rotary_policy": "checkpoint"
+    }
+  ],
+  "admission": "Selected by Jason by listening; native conversion receipt guitar-quality-native.json. Redistribution status unresolved."
+}

Guitar/guitar-quality-native.json

@@ -0,0 +1,17 @@
+{
+  "mode": "quality",
+  "candidate": "SW",
+  "file": "guitar-quality.onnx",
+  "sha256": "667116d92813526e704c908574e44e9d0c4cd21fb44337ce5a68fb93df0b2f39",
+  "bytes": 207712098,
+  "chunk": 588800,
+  "hop": 512,
+  "fft": 2048,
+  "overlap": 2,
+  "times": 1151,
+  "attention_ops": 24,
+  "linear_ops": 306,
+  "attention_parity_max": 0.0,
+  "rotary_policy": "checkpoint",
+  "scope": "User-authorized local development integration; commercial redistribution not reviewed."
+}

Guitar/guitar-quality.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:667116d92813526e704c908574e44e9d0c4cd21fb44337ce5a68fb93df0b2f39
+size 207712098

Hybrid/LICENSE-MSS-Code.txt

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2024 Roman Solovyev (ZFTurbo)
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

Hybrid/LICENSE-MelBand-Drums.txt

@@ -0,0 +1,188 @@
+Mel-Band RoFormer four-stem Large v1 — drum model weights
+
+Unravel's Hybrid drum model (mel-drums.onnx) is a conversion by StrumWorks of the
+"Mel-Band RoFormer four-stem Large v1" weights by Aname-Tommy, published at
+https://huggingface.co/Aname-Tommy/melbandroformer4stems
+(revision 78c9bfdc3ac4122673fb2b6b19a6924e8e89e3a4) under the Apache License 2.0, as
+declared in the author's model metadata. The conversion changes the format only (ONNX,
+drum head); intake receipt: mel-native.json. The model architecture code is used under
+the MIT licence in LICENSE-MSS-Code.txt.
+
+The Apache License 2.0 follows.
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+    TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+    1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+    2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+    3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+    4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+    5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+    6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+    7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+    8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+    9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+    END OF TERMS AND CONDITIONS

Hybrid/LICENSE-SCNet-Code.txt

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2024 Roman Solovyev (ZFTurbo)
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

Hybrid/mel-drums.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a53c1240d2ee96c13776a5abc40f434a351f8a5f19077df0d1d134948d176c6
+size 1107389260

Hybrid/mel-native.json

@@ -0,0 +1,98 @@
+{
+  "kind": "mel",
+  "source": {
+    "candidate": "Aname Mel-Band RoFormer four-stem Large v1",
+    "scope": "Local development evaluation; not installed as app default",
+    "source_revision": "6fdad13686bb7e84d029ec66f9dbe76e65c93ebf",
+    "weights_revision": "78c9bfdc3ac4122673fb2b6b19a6924e8e89e3a4",
+    "weights_author": "Aname-Tommy",
+    "author_license_metadata": "apache-2.0",
+    "roles": [
+      "drums",
+      "bass",
+      "other",
+      "vocals"
+    ],
+    "sample_rate": 44100,
+    "chunk_frames": 661500,
+    "overlap": 4,
+    "batch_size": 1,
+    "precision": "float32",
+    "sha256": {
+      "Models/Evaluation/MelBandAname/model.ckpt": "a2ca1ac95e203d2630994c3d4a0cebabd7160e70bb010adab794843340fb704f",
+      "Models/Evaluation/MelBandAname/config.yaml": "65eaffebffcf09d496e35ee9340a9089e869aebdb30b203f883c2c9dbda6b0b0",
+      "Models/Evaluation/MelBandAname/README.md": "4bcf87ecfbbb8e07a01b21415a970c8b53a5283bf6872b657040d3f45c9241f7",
+      "Development/RoFormer/source/models/bs_roformer/mel_band_roformer.py": "a28fce62020faad6461176660e012cf5799ae8a26c76fe4b2bc636f5b44b9fc5",
+      "Development/RoFormer/source/models/bs_roformer/attend.py": "c7abbc40a3fd20ff7f6001fa9f8ee9ad5df6452272712a5e719b8f8468bf2223",
+      "Development/RoFormer/source/LICENSE": "3282dc057695ef5b9a64909a7092ca40b2c292c232580fc6ace6e5d665cc0207"
+    },
+    "downloads": [
+      {
+        "path": "Models/Evaluation/MelBandAname/model.ckpt",
+        "url": "https://huggingface.co/Aname-Tommy/melbandroformer4stems/resolve/78c9bfdc3ac4122673fb2b6b19a6924e8e89e3a4/mel_band_roformer_4stems_large_ver1.ckpt",
+        "bytes": 3764030874
+      },
+      {
+        "path": "Models/Evaluation/MelBandAname/config.yaml",
+        "url": "https://huggingface.co/Aname-Tommy/melbandroformer4stems/resolve/78c9bfdc3ac4122673fb2b6b19a6924e8e89e3a4/config_large.yaml",
+        "bytes": 4606
+      },
+      {
+        "path": "Models/Evaluation/MelBandAname/README.md",
+        "url": "https://huggingface.co/Aname-Tommy/melbandroformer4stems/resolve/78c9bfdc3ac4122673fb2b6b19a6924e8e89e3a4/README.md",
+        "bytes": 31
+      },
+      {
+        "path": "Development/RoFormer/source/models/bs_roformer/mel_band_roformer.py",
+        "url": "https://raw.githubusercontent.com/ZFTurbo/Music-Source-Separation-Training/6fdad13686bb7e84d029ec66f9dbe76e65c93ebf/models/bs_roformer/mel_band_roformer.py",
+        "bytes": 22708
+      },
+      {
+        "path": "Development/RoFormer/source/models/bs_roformer/attend.py",
+        "url": "https://raw.githubusercontent.com/ZFTurbo/Music-Source-Separation-Training/6fdad13686bb7e84d029ec66f9dbe76e65c93ebf/models/bs_roformer/attend.py",
+        "bytes": 4410
+      },
+      {
+        "path": "Development/RoFormer/source/LICENSE",
+        "url": "https://raw.githubusercontent.com/ZFTurbo/Music-Source-Separation-Training/6fdad13686bb7e84d029ec66f9dbe76e65c93ebf/LICENSE",
+        "bytes": 1081
+      }
+    ]
+  },
+  "input_shape": [
+    1,
+    4,
+    2049,
+    751
+  ],
+  "chunk_frames": 661500,
+  "transform_frames": 661500,
+  "fft": 4096,
+  "hop": 882,
+  "normalized": false,
+  "overlap": 4,
+  "waveform_parity": {
+    "relative_rms": 6.018953513375891e-08,
+    "max_absolute": 1.7881393432617188e-07
+  },
+  "file": "mel-drums.onnx",
+  "sha256": "5a53c1240d2ee96c13776a5abc40f434a351f8a5f19077df0d1d134948d176c6",
+  "bytes": 1107389260,
+  "checks": [
+    {
+      "scale": 1.0,
+      "relative_rms": 9.059534630569486e-07,
+      "max_absolute": 0.00054931640625,
+      "waveform_relative_rms": 4.748359256866445e-07,
+      "waveform_max_absolute": 6.556510925292969e-07
+    },
+    {
+      "scale": 0.001,
+      "relative_rms": 8.877570036715101e-07,
+      "max_absolute": 5.364418029785156e-07,
+      "waveform_relative_rms": 4.708283375891488e-07,
+      "waveform_max_absolute": 6.402842700481415e-10
+    }
+  ],
+  "exporter_sha256": "0336edd03b31e141084a965fe6e200fac894e2416052ccb544678143958d97c7"
+}

Hybrid/scnet-bass.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fc87794d8fc78fbadc8ca499ff385331a444ea35928cfa165b89db18065b7c8
+size 170947180

Hybrid/scnet-native.json

@@ -0,0 +1,124 @@
+{
+  "kind": "scnet",
+  "source": {
+    "candidate": "SCNet Large starrytong fixed, release v1.0.9",
+    "scope": "Local bass evaluation; not an installed or shipping model",
+    "source_revision": "6fdad13686bb7e84d029ec66f9dbe76e65c93ebf",
+    "roles": [
+      "drums",
+      "bass",
+      "other",
+      "vocals"
+    ],
+    "sample_rate": 44100,
+    "chunk_frames": 485100,
+    "overlap": 4,
+    "batch_size": 1,
+    "device": "cpu",
+    "precision": "float32",
+    "input_scaling": "Upstream mono mean/std normalization, reversed per output; no independent output peak normalization",
+    "code_license": "MSST MIT license retained",
+    "weights_status": "Public author-linked checkpoint; redistribution admission remains open",
+    "hash_provenance": "SHA-256 computed from official HTTPS release/source downloads; release API does not publish digests for these assets",
+    "sha256": {
+      "Models/Evaluation/SCNetLarge/model.ckpt": "65900dfa07d6b6e5d784c0f143920200a4bd281d6e78a806c549d0b912d5885e",
+      "Models/Evaluation/SCNetLarge/config.yaml": "3efa43af7c50d4a6e1105d2599d2d6a31c8012ee99eaad9a7e80e057945d44ee",
+      "Development/SCNet/source/scnet/__init__.py": "8ef8829a84f72fb500ee364c0394527cbbba7f86cd04f76052bea5824e5b9a12",
+      "Development/SCNet/source/scnet/scnet.py": "a024c6a6b3e5eda6da29f052076772c277183dc273f144b4feca1c565030e1ed",
+      "Development/SCNet/source/scnet/separation.py": "4fdda9982003b795c3b5d964f998ca0bc416d7c232b813706a6088b16a1a1639",
+      "Development/SCNet/source/LICENSE": "3282dc057695ef5b9a64909a7092ca40b2c292c232580fc6ace6e5d665cc0207",
+      "Development/SCNet/reference/inference.py": "77401006b5eaebf2ce3e0770a5fdb4bfabbce8def17e40ba2c36a592e71522c7",
+      "Development/SCNet/reference/utils/audio_utils.py": "70721f4fc26467ecec21557c842333747ed4c765d0a842f9ae22799280f5970a",
+      "Development/SCNet/reference/utils/model_utils.py": "27d87727cf23a8a091417535c748f9b4692c7b3ddb3de405db5ccde22a5512d1",
+      "Development/SCNet/reference/utils/settings.py": "eafcec0edf8e624103efa2efe572a45c308ca1e8cb621291f31b7ea89316ea81"
+    },
+    "downloads": [
+      {
+        "path": "Models/Evaluation/SCNetLarge/model.ckpt",
+        "url": "https://github.com/ZFTurbo/Music-Source-Separation-Training/releases/download/v1.0.9/SCNet-large_starrytong_fixed.ckpt",
+        "bytes": 168852258
+      },
+      {
+        "path": "Models/Evaluation/SCNetLarge/config.yaml",
+        "url": "https://github.com/ZFTurbo/Music-Source-Separation-Training/releases/download/v1.0.9/config_musdb18_scnet_large_starrytong.yaml",
+        "bytes": 1955
+      },
+      {
+        "path": "Development/SCNet/source/scnet/__init__.py",
+        "url": "https://raw.githubusercontent.com/ZFTurbo/Music-Source-Separation-Training/6fdad13686bb7e84d029ec66f9dbe76e65c93ebf/models/scnet/__init__.py",
+        "bytes": 25
+      },
+      {
+        "path": "Development/SCNet/source/scnet/scnet.py",
+        "url": "https://raw.githubusercontent.com/ZFTurbo/Music-Source-Separation-Training/6fdad13686bb7e84d029ec66f9dbe76e65c93ebf/models/scnet/scnet.py",
+        "bytes": 13750
+      },
+      {
+        "path": "Development/SCNet/source/scnet/separation.py",
+        "url": "https://raw.githubusercontent.com/ZFTurbo/Music-Source-Separation-Training/6fdad13686bb7e84d029ec66f9dbe76e65c93ebf/models/scnet/separation.py",
+        "bytes": 3742
+      },
+      {
+        "path": "Development/SCNet/source/LICENSE",
+        "url": "https://raw.githubusercontent.com/ZFTurbo/Music-Source-Separation-Training/6fdad13686bb7e84d029ec66f9dbe76e65c93ebf/LICENSE",
+        "bytes": 1081
+      },
+      {
+        "path": "Development/SCNet/reference/inference.py",
+        "url": "https://raw.githubusercontent.com/ZFTurbo/Music-Source-Separation-Training/6fdad13686bb7e84d029ec66f9dbe76e65c93ebf/inference.py",
+        "bytes": 8366
+      },
+      {
+        "path": "Development/SCNet/reference/utils/audio_utils.py",
+        "url": "https://raw.githubusercontent.com/ZFTurbo/Music-Source-Separation-Training/6fdad13686bb7e84d029ec66f9dbe76e65c93ebf/utils/audio_utils.py",
+        "bytes": 12758
+      },
+      {
+        "path": "Development/SCNet/reference/utils/model_utils.py",
+        "url": "https://raw.githubusercontent.com/ZFTurbo/Music-Source-Separation-Training/6fdad13686bb7e84d029ec66f9dbe76e65c93ebf/utils/model_utils.py",
+        "bytes": 39406
+      },
+      {
+        "path": "Development/SCNet/reference/utils/settings.py",
+        "url": "https://raw.githubusercontent.com/ZFTurbo/Music-Source-Separation-Training/6fdad13686bb7e84d029ec66f9dbe76e65c93ebf/utils/settings.py",
+        "bytes": 32962
+      }
+    ]
+  },
+  "input_shape": [
+    1,
+    4,
+    2049,
+    476
+  ],
+  "chunk_frames": 485100,
+  "transform_frames": 486400,
+  "fft": 4096,
+  "hop": 1024,
+  "normalized": true,
+  "overlap": 4,
+  "waveform_parity": {
+    "relative_rms": 4.2188591464764613e-07,
+    "max_absolute": 1.0132789611816406e-06
+  },
+  "file": "scnet-bass.onnx",
+  "sha256": "5fc87794d8fc78fbadc8ca499ff385331a444ea35928cfa165b89db18065b7c8",
+  "bytes": 170947180,
+  "checks": [
+    {
+      "scale": 1.0,
+      "relative_rms": 0.0008372841177720371,
+      "max_absolute": 0.0639657974243164,
+      "waveform_relative_rms": 0.0008056081704676091,
+      "waveform_max_absolute": 0.0026332437992095947
+    },
+    {
+      "scale": 0.001,
+      "relative_rms": 0.0018604112198468226,
+      "max_absolute": 0.00025827065110206604,
+      "waveform_relative_rms": 0.0018673858670629061,
+      "waveform_max_absolute": 4.628859460353851e-05
+    }
+  ],
+  "exporter_sha256": "0336edd03b31e141084a965fe6e200fac894e2416052ccb544678143958d97c7"
+}

LICENSE.md

@@ -0,0 +1,67 @@
+# Unravel runtime models: licences and provenance
+
+This repository holds model weights from different authors under different terms. There is
+no single licence. StrumWorks' own conversion code and receipts do not change the terms of the
+underlying weights. Status values are stated as they stand, and none is described as cleared
+unless it is:
+
+- **MIT / Apache-2.0**: the weight author published that licence.
+- **research-only**: the author limited the weights to research or scientific use.
+- **unresolved**: no licence or permission covering the weights has been established.
+
+Each model folder contains its receipt (source URLs, revisions and SHA-256 of the original
+download) and the code licences that apply.
+
+## HTDemucs 6-source (MLX conversion)
+
+- **Status:** research-only
+- **Author:** Alexandre Défossez / Meta AI (Demucs v4); MLX conversion by iky1e
+- **Source:** https://huggingface.co/iky1e/demucs-mlx @ 3ff2e582c1456c6a03a077b6e47254e0ba41bdcf
+- **Used for:** import-fast
+- **Receipt:** `Fast/htdemucs-mlx.json` (inside the app)
+- **Notes:** The Demucs author described the supplied pretrained weights as outside the MIT code licence and intended for scientific use. The MLX repository is labelled MIT, which is not the weight author’s grant.
+
+## MDX23C DrumSep six-part drum model (ONNX conversion)
+
+- **Status:** unresolved
+- **Author:** aufr33 and jarredou; ONNX conversion by StrumWorks
+- **Source:** Public model repository, release weights-drumsep-v1 (checkpoint SHA-256 d2a4aa53eb584d21eead358a4e66d1882ad182911be018f052b5da73be9096d0); inference code revision cb4779eedeca4f55ba498df8e234e2b5f4b90aaf
+- **Used for:** split-drums-fast, split-drums-quality
+- **Receipt:** `Evaluation/mdx23c-native.json`
+- **Notes:** The original checkpoint’s terms are undocumented. MIT wrapper and mirror labels do not establish permission for the weights. Code: MDX23C MIT (LICENSE-MDX23C-Code.txt).
+
+## Mel-Band RoFormer four-stem Large v1, drums (ONNX conversion)
+
+- **Status:** Apache-2.0
+- **Author:** Aname-Tommy; ONNX conversion by StrumWorks
+- **Source:** https://huggingface.co/Aname-Tommy/melbandroformer4stems @ 78c9bfdc3ac4122673fb2b6b19a6924e8e89e3a4
+- **Used for:** split-drums-quality
+- **Receipt:** `Hybrid/mel-native.json`
+- **Notes:** Licence declared in the author’s Hugging Face model metadata (apache-2.0). Apache-2.0 notice duties apply. Code: Music-Source-Separation-Training MIT.
+
+## SCNet Large (starrytong, fixed), bass (ONNX conversion)
+
+- **Status:** unresolved
+- **Author:** starrytong (SCNet); checkpoint released by ZFTurbo, MSST v1.0.9; ONNX conversion by StrumWorks
+- **Source:** https://github.com/ZFTurbo/Music-Source-Separation-Training/releases/tag/v1.0.9 (code 6fdad13686bb7e84d029ec66f9dbe76e65c93ebf)
+- **Used for:** enhance-bass
+- **Receipt:** `Hybrid/scnet-native.json`
+- **Notes:** Public, author-linked checkpoint; permission to redistribute the weights has not been established. Code: MIT.
+
+## BS-RoFormer SW (fixed), guitar (ONNX conversion)
+
+- **Status:** unresolved
+- **Author:** enerjazzer (BS-ROFO-SW-Fixed); ONNX conversion by StrumWorks
+- **Source:** https://huggingface.co/enerjazzer/BS-ROFO-SW-Fixed @ a443a2985534b3bc815ef54a5d446c6a0390f974
+- **Used for:** enhance-guitar
+- **Receipt:** `Guitar/guitar-evaluation.json`
+- **Notes:** No licence is published for the weights. Code: Music-Source-Separation-Training MIT.
+
+## Kim Mel-Band RoFormer vocals (ONNX conversion)
+
+- **Status:** MIT
+- **Author:** Kimberley Jensen; ONNX conversion by StrumWorks
+- **Source:** https://huggingface.co/KimberleyJSN/melbandroformer @ ac9b0614ab3cd7f77219e18ba494dfd93956c348; code github.com/KimberleyJensen/Mel-Band-Roformer-Vocal-Model @ 25f44ffb55ee3c301281bba21b2d6d311cb69ae2
+- **Used for:** enhance-vocals
+- **Receipt:** `Vocals/kim-vocals-evaluation.json`
+- **Notes:** The model card declares MIT, and the author confirmed MIT for her models (issue #18).

README.md

@@ -0,0 +1,46 @@
+---
+license: other
+license_name: unravel-mixed-licenses
+license_link: LICENSE.md
+tags:
+- audio-source-separation
+- unravel
+---
+
+# Unravel runtime models (StrumWorks)
+
+The separation models that **Unravel by StrumWorks** downloads, laid out exactly like
+the app's model folder, and the signed catalog that lists them.
+
+- `catalog.json` is signed with StrumWorks' Ed25519 key. Unravel checks the signature
+  against keys built into the app before reading it.
+- The catalog pins every download to an exact commit of this repository, and Unravel
+  verifies each file's size and SHA-256 before using it.
+- Separation runs entirely on the user's Mac; nothing is uploaded.
+
+**Licences differ per model and not every model is cleared for redistribution.**
+Status values used: Apache-2.0, MIT, research-only, unresolved. See [LICENSE.md](LICENSE.md) for each model's
+author, source revision, receipt and status.
+
+| File | Bytes | SHA-256 | Model | Licence status |
+|---|---:|---|---|---|
+| `Fast/Models/htdemucs_6s.safetensors` | 109,726,583 | `d298f7f746bf53c21baad44fb08e88807ef47feb551dd22f1601a546c85b8e02` | HTDemucs 6-source (MLX conversion) | research-only |
+| `Fast/Models/htdemucs_6s_config.json` | 1,946 | `97f8315891d8edc9aa6f59e56e0d352fbad5ebfb8a4faf46341ab2f1844596a9` | HTDemucs 6-source (MLX conversion) | research-only |
+| `Evaluation/LICENSE-MDX23C-Code.txt` | 1,679 | `bf7f9317b1716d7afa02784949f25539c811c2aefcd613867a372cc8f52dc7c3` | MDX23C DrumSep six-part drum model (ONNX conversion) | unresolved |
+| `Evaluation/mdx23c-drums.onnx` | 437,702,629 | `99d68124a0eabb3291bf155334fb55ee0af1bc9957d54bd2167a7a3458aa7070` | MDX23C DrumSep six-part drum model (ONNX conversion) | unresolved |
+| `Evaluation/mdx23c-native.json` | 1,334 | `0bb0650456b509a6c2ef55369fd291e15e6ae8d2308a3aee50a72b1bcb405bc0` | MDX23C DrumSep six-part drum model (ONNX conversion) | unresolved |
+| `Hybrid/LICENSE-MSS-Code.txt` | 1,081 | `3282dc057695ef5b9a64909a7092ca40b2c292c232580fc6ace6e5d665cc0207` | Mel-Band RoFormer four-stem Large v1, drums (ONNX conversion) | Apache-2.0 |
+| `Hybrid/LICENSE-MelBand-Drums.txt` | 10,794 | `35fbcfa0ed111bceb5d4ab5fa63e0aa6d1c5ce6e4bb026f9686eaadceb249e00` | Mel-Band RoFormer four-stem Large v1, drums (ONNX conversion) | Apache-2.0 |
+| `Hybrid/mel-drums.onnx` | 1,107,389,260 | `5a53c1240d2ee96c13776a5abc40f434a351f8a5f19077df0d1d134948d176c6` | Mel-Band RoFormer four-stem Large v1, drums (ONNX conversion) | Apache-2.0 |
+| `Hybrid/mel-native.json` | 3,940 | `ee61a8e7e1debc49a873f8615cd8ce2b4fa01b6837173acd15f9ad4794a6bbc1` | Mel-Band RoFormer four-stem Large v1, drums (ONNX conversion) | Apache-2.0 |
+| `Hybrid/LICENSE-SCNet-Code.txt` | 1,081 | `3282dc057695ef5b9a64909a7092ca40b2c292c232580fc6ace6e5d665cc0207` | SCNet Large (starrytong, fixed), bass (ONNX conversion) | unresolved |
+| `Hybrid/scnet-bass.onnx` | 170,947,180 | `5fc87794d8fc78fbadc8ca499ff385331a444ea35928cfa165b89db18065b7c8` | SCNet Large (starrytong, fixed), bass (ONNX conversion) | unresolved |
+| `Hybrid/scnet-native.json` | 5,642 | `95d3112e7447f3423cce3bdd7b65a6df52187dd9460308049eed7e9dd2740b48` | SCNet Large (starrytong, fixed), bass (ONNX conversion) | unresolved |
+| `Guitar/LICENSE-MSS-Code.txt` | 1,081 | `3282dc057695ef5b9a64909a7092ca40b2c292c232580fc6ace6e5d665cc0207` | BS-RoFormer SW (fixed), guitar (ONNX conversion) | unresolved |
+| `Guitar/guitar-evaluation.json` | 2,270 | `ab9e6ca95ea53cc6d7ab1c91f15acd1c5f1722b291428de72fbc246919c497c8` | BS-RoFormer SW (fixed), guitar (ONNX conversion) | unresolved |
+| `Guitar/guitar-quality-native.json` | 470 | `5a6963665cba73fbec24eab9d82eb80d6eecb29a1499a4b31c21f020fcedc855` | BS-RoFormer SW (fixed), guitar (ONNX conversion) | unresolved |
+| `Guitar/guitar-quality.onnx` | 207,712,098 | `667116d92813526e704c908574e44e9d0c4cd21fb44337ce5a68fb93df0b2f39` | BS-RoFormer SW (fixed), guitar (ONNX conversion) | unresolved |
+| `Vocals/LICENSE-Kim-Vocals.txt` | 1,611 | `66f9995ade40db1de98b5e2f78a51c7ccffb0f4881bd87ec7c4f92fb71ea44cb` | Kim Mel-Band RoFormer vocals (ONNX conversion) | MIT |
+| `Vocals/kim-vocals-evaluation.json` | 5,439 | `2e93547d2088585873205f075423f6d6456f275c6b381d93c47980c2bf341bc8` | Kim Mel-Band RoFormer vocals (ONNX conversion) | MIT |
+| `Vocals/vocals-kim.onnx` | 929,362,776 | `94897c6d3616f3a98e5fbf18b12c9515ce4bfc8fd0707da701deae67b4c6c758` | Kim Mel-Band RoFormer vocals (ONNX conversion) | MIT |
+| `Vocals/vocals-native.json` | 871 | `4df8c13afc0db4d5dffa61dd4bfb5306db3a96c89517786c1bcd8b93d98c730e` | Kim Mel-Band RoFormer vocals (ONNX conversion) | MIT |

Vocals/LICENSE-Kim-Vocals.txt

@@ -0,0 +1,25 @@
+Kim Mel-Band RoFormer vocal model (MelBandRoformer.ckpt)
+Author: KimberleyJSN (Kimberley Jensen)
+Weights: https://huggingface.co/KimberleyJSN/melbandroformer (revision ac9b0614ab3cd7f77219e18ba494dfd93956c348)
+Code: https://github.com/KimberleyJensen/Mel-Band-Roformer-Vocal-Model (commit 25f44ffb55ee3c301281bba21b2d6d311cb69ae2)
+
+Licence: MIT. The model card declares `license: mit`, and the author confirmed
+MIT for her models (https://github.com/KimberleyJensen/Mel-Band-Roformer-Vocal-Model/issues/18).
+The repository itself contains no LICENSE file; the standard MIT terms follow.
+
+Permission is hereby granted, free of charge, to any person obtaining a copy of
+this software and associated documentation files (the "Software"), to deal in
+the Software without restriction, including without limitation the rights to
+use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of
+the Software, and to permit persons to whom the Software is furnished to do so,
+subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS
+FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR
+COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN
+AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Vocals/kim-vocals-evaluation.json

@@ -0,0 +1,106 @@
+{
+  "kind": "vocals",
+  "candidate": "Kim Mel-Band RoFormer vocal model (MelBandRoformer.ckpt)",
+  "selection": "Selected directly by Jason (24 October 2026); no blind comparison.",
+  "weights_repository": "https://huggingface.co/KimberleyJSN/melbandroformer",
+  "weights_revision": "ac9b0614ab3cd7f77219e18ba494dfd93956c348",
+  "source_repository": "https://github.com/KimberleyJensen/Mel-Band-Roformer-Vocal-Model",
+  "source_revision": "25f44ffb55ee3c301281bba21b2d6d311cb69ae2",
+  "license": {
+    "statement": "MIT. The Hugging Face model card at the pinned revision declares `license: mit` in its front matter, and the author confirmed MIT for her models (see MODEL_RESEARCH.md, Kim author confirmation, issue #18).",
+    "evidence": [
+      "https://huggingface.co/KimberleyJSN/melbandroformer",
+      "https://github.com/KimberleyJensen/Mel-Band-Roformer-Vocal-Model/issues/18"
+    ],
+    "note": "The GitHub repository at the pinned commit contains no LICENSE file; the model card and author statement are the licence evidence."
+  },
+  "load": {
+    "api": "torch.load(weights_only=True) + load_state_dict(strict=True)",
+    "tensors": 684,
+    "parameters": 228202852,
+    "dtype": "float32",
+    "tensor_bytes": 912812688,
+    "largest_group": "mask_estimators (805.9 MB)"
+  },
+  "config": {
+    "dim": 384,
+    "depth": 6,
+    "time_transformer_depth": 1,
+    "freq_transformer_depth": 1,
+    "num_bands": 60,
+    "heads": 8,
+    "dim_head": 64,
+    "mask_estimator_depth": 2,
+    "sample_rate": 44100,
+    "stft_n_fft": 2048,
+    "stft_hop_length": 441,
+    "stft_win_length": 2048,
+    "stft_normalized": false,
+    "chunk_size": 352800,
+    "num_overlap": 2,
+    "target_instrument": "vocals"
+  },
+  "scope": "Source assets stay under Models/Evaluation/ (not in Git). Converted native graph is described in vocals-native.json.",
+  "downloads": [
+    {
+      "path": "Models/Evaluation/KimMelBandVocals/MelBandRoformer.ckpt",
+      "url": "https://huggingface.co/KimberleyJSN/melbandroformer/resolve/ac9b0614ab3cd7f77219e18ba494dfd93956c348/MelBandRoformer.ckpt",
+      "bytes": 913106900,
+      "sha256": "87201f4d31afb5bc79993230fc49446918425574db48c01c405e44f365c7559e"
+    },
+    {
+      "path": "Models/Evaluation/KimMelBandVocals/README-model-card.md",
+      "url": "https://huggingface.co/KimberleyJSN/melbandroformer/resolve/ac9b0614ab3cd7f77219e18ba494dfd93956c348/README.md",
+      "bytes": 20,
+      "sha256": "5c1b3fd193f37b9dcdd947a12e566189f1c9d2664f4e5e95548e6a4b9fc7adab"
+    },
+    {
+      "path": "Models/Evaluation/KimMelBandVocals/source/configs/config_vocals_mel_band_roformer.yaml",
+      "url": "https://raw.githubusercontent.com/KimberleyJensen/Mel-Band-Roformer-Vocal-Model/25f44ffb55ee3c301281bba21b2d6d311cb69ae2/configs/config_vocals_mel_band_roformer.yaml",
+      "bytes": 733,
+      "sha256": "87aabb300193b019159269b69d6fe5f313aae4201a22b4af6268bc2846ab2fe1"
+    },
+    {
+      "path": "Models/Evaluation/KimMelBandVocals/source/models/mel_band_roformer/__init__.py",
+      "url": "https://raw.githubusercontent.com/KimberleyJensen/Mel-Band-Roformer-Vocal-Model/25f44ffb55ee3c301281bba21b2d6d311cb69ae2/models/mel_band_roformer/__init__.py",
+      "bytes": 71,
+      "sha256": "bde4b0739ca79d0d85ee9cf1df6eea0558c8e17e06414fbb3600897ff6d24640"
+    },
+    {
+      "path": "Models/Evaluation/KimMelBandVocals/source/models/mel_band_roformer/attend.py",
+      "url": "https://raw.githubusercontent.com/KimberleyJensen/Mel-Band-Roformer-Vocal-Model/25f44ffb55ee3c301281bba21b2d6d311cb69ae2/models/mel_band_roformer/attend.py",
+      "bytes": 2854,
+      "sha256": "9ffd7d06b22a53bb8e6162adf674e7298d2227e1eea4f6404cb89ff1bbefecbe"
+    },
+    {
+      "path": "Models/Evaluation/KimMelBandVocals/source/models/mel_band_roformer/mel_band_roformer.py",
+      "url": "https://raw.githubusercontent.com/KimberleyJensen/Mel-Band-Roformer-Vocal-Model/25f44ffb55ee3c301281bba21b2d6d311cb69ae2/models/mel_band_roformer/mel_band_roformer.py",
+      "bytes": 16726,
+      "sha256": "d0e958c58114872c3a0af54f97719e07690ff5d98473fcd8b48485f776737b8f"
+    },
+    {
+      "path": "Models/Evaluation/KimMelBandVocals/source/inference.py",
+      "url": "https://raw.githubusercontent.com/KimberleyJensen/Mel-Band-Roformer-Vocal-Model/25f44ffb55ee3c301281bba21b2d6d311cb69ae2/inference.py",
+      "bytes": 4532,
+      "sha256": "3d86bc364497a46d9a154ee87a1a64ea9d4e7b3036402ede8ac9dd454842a9c9"
+    },
+    {
+      "path": "Models/Evaluation/KimMelBandVocals/source/utils.py",
+      "url": "https://raw.githubusercontent.com/KimberleyJensen/Mel-Band-Roformer-Vocal-Model/25f44ffb55ee3c301281bba21b2d6d311cb69ae2/utils.py",
+      "bytes": 4161,
+      "sha256": "c68f49bb593347e256f5001dea1f380e9c575c331afc6b2920e8cdbdcd188be2"
+    },
+    {
+      "path": "Models/Evaluation/KimMelBandVocals/source/README.md",
+      "url": "https://raw.githubusercontent.com/KimberleyJensen/Mel-Band-Roformer-Vocal-Model/25f44ffb55ee3c301281bba21b2d6d311cb69ae2/README.md",
+      "bytes": 2320,
+      "sha256": "b5bd4ad05f3a861c4ac1c19f404385ad4152ad1e4c5ae4acca099a488c44da7f"
+    },
+    {
+      "path": "Models/Evaluation/KimMelBandVocals/source/requirements.txt",
+      "url": "https://raw.githubusercontent.com/KimberleyJensen/Mel-Band-Roformer-Vocal-Model/25f44ffb55ee3c301281bba21b2d6d311cb69ae2/requirements.txt",
+      "bytes": 189,
+      "sha256": "f2120602e3d17d9722c443d87a541f52cbecd2af94406247bf49aaabae403a15"
+    }
+  ]
+}

Vocals/vocals-kim.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94897c6d3616f3a98e5fbf18b12c9515ce4bfc8fd0707da701deae67b4c6c758
+size 929362776

Vocals/vocals-native.json

@@ -0,0 +1,23 @@
+{
+  "kind": "vocals",
+  "candidate": "Kim Mel-Band RoFormer",
+  "file": "vocals-kim.onnx",
+  "sha256": "94897c6d3616f3a98e5fbf18b12c9515ce4bfc8fd0707da701deae67b4c6c758",
+  "bytes": 929362776,
+  "chunk": 352800,
+  "hop": 441,
+  "fft": 2048,
+  "overlap": 2,
+  "times": 801,
+  "attention_ops": 12,
+  "linear_ops": 300,
+  "attention_parity_max": 3.3155083656311035e-07,
+  "waveform_boundary_parity": {
+    "relative_rms": 5.970930061494073e-08,
+    "max_absolute": 1.7881393432617188e-07
+  },
+  "source_receipt_sha256": "2e93547d2088585873205f075423f6d6456f275c6b381d93c47980c2bf341bc8",
+  "exporter_sha256": "f0951fe48acb58224de86decdf1e7b8520636691416328f1c83accc4f908d873",
+  "rotary_embedding_torch": "rotary_embedding_torch from KIM_PINS",
+  "scope": "Selected by Jason; MIT per the model card and author confirmation. Weights install to the models folder, not Git."
+}