feat: add pc-ddsp, bigvgan and refactor a lot

.pre-commit-config.yaml

@@ -53,11 +53,11 @@ repos:
     rev: 23.3.0
     hooks:
       - id: black
-  - repo: https://github.com/codespell-project/codespell
-    rev: v2.2.4
-    hooks:
-      - id: codespell
-        args: [-w]
+  #- repo: https://github.com/codespell-project/codespell
+  #  rev: v2.2.4
+  #  hooks:
+  #    - id: codespell
+  #      args: [-w]
   - repo: https://github.com/PyCQA/flake8
     rev: 6.0.0
     hooks:

LICENSE

@@ -1,6 +1,10 @@
 MIT License
 
 Copyright (c) 2023 34j and contributors
+Copyright (c) 2022 NVIDIA CORPORATION.
+Copyright (c) 2020 Edward Dixon
+Copyright 2020 Alexandre Défossez
+Copyright (c) 2023 yxlllc
 Copyright (c) 2021 Jingyi Li
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
@@ -221,3 +225,32 @@ SOFTWARE.
    WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
    See the License for the specific language governing permissions and
    limitations under the License.
+BSD 3-Clause License
+
+Copyright (c) 2019, Seungwon Park 박승원
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+1. Redistributions of source code must retain the above copyright notice, this
+   list of conditions and the following disclaimer.
+
+2. Redistributions in binary form must reproduce the above copyright notice,
+   this list of conditions and the following disclaimer in the documentation
+   and/or other materials provided with the distribution.
+
+3. Neither the name of the copyright holder nor the names of its
+   contributors may be used to endorse or promote products derived from
+   this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

pyproject.toml

@@ -78,6 +78,16 @@ myst-parser = ">=0.16"
 sphinx = ">=4.0"
 sphinx-rtd-theme = ">=1.0"
 
+
+[tool.poetry.group.pc_ddsp.dependencies]
+gin = "^0.1.6"
+gin-config = "^0.5.0"
+local-attention = "^1.8.5"
+
+
+[tool.poetry.group.bigvgan.dependencies]
+alias-free-torch = "^0.0.6"
+
 [tool.semantic_release]
 branch = "main"
 version_toml = "pyproject.toml:tool.poetry.version"

src/so_vits_svc_fork/dataset.py

@@ -64,7 +64,7 @@ def _pad_stack(array: Sequence[torch.Tensor]) -> torch.Tensor:
 class TextAudioCollate(nn.Module):
     def forward(
         self, batch: Sequence[dict[str, torch.Tensor]]
-    ) -> tuple[torch.Tensor, ...]:
+    ) -> dict[str, torch.Tensor]:
         batch = [b for b in batch if b is not None]
         batch = list(sorted(batch, key=lambda x: x["mel_spec"].shape[1], reverse=True))
         lengths = torch.tensor([b["mel_spec"].shape[1] for b in batch]).long()
@@ -74,14 +74,5 @@ def forward(
                 results[key] = _pad_stack([b[key] for b in batch]).cpu()
             else:
                 results[key] = torch.tensor([[b[key]] for b in batch]).cpu()
-
-        return (
-            results["content"],
-            results["f0"],
-            results["spec"],
-            results["mel_spec"],
-            results["audio"],
-            results["spk"],
-            lengths,
-            results["uv"],
-        )
+        results["length"] = lengths
+        return results

src/so_vits_svc_fork/f0.py

@@ -8,16 +8,16 @@
 import torchcrepe
 from cm_time import timer
 from numpy import dtype, float32, ndarray
-from torch import FloatTensor, Tensor
+from torch import Tensor
 
 from so_vits_svc_fork.utils import get_optimal_device
 
 LOG = getLogger(__name__)
 
 
 def normalize_f0(
-    f0: FloatTensor, x_mask: FloatTensor, uv: FloatTensor, random_scale=True
-) -> FloatTensor:
+    f0: Tensor, x_mask: Tensor, uv: Tensor, random_scale: bool = True
+) -> Tensor:
     # calculate means based on x_mask
     uv_sum = torch.sum(uv, dim=1, keepdim=True)
     uv_sum[uv_sum == 0] = 9999

src/so_vits_svc_fork/inference/core.py

@@ -216,10 +216,10 @@ def infer(
                 audio = self.net_g.infer(
                     c,
                     f0=f0,
-                    g=sid,
+                    spk=sid,
                     uv=uv,
                     predict_f0=auto_predict_f0,
-                    noice_scale=noise_scale,
+                    noise_scale=noise_scale,
                 )[0, 0].data.float()
             audio_duration = audio.shape[-1] / self.target_sample
             LOG.info(