fix: change f0 method from crepe to dio

BREAKING CHANGE: all default values for f0 method changed
voicepaw · Mar 24, 2023 · 24353e6 · 24353e6
1 parent 59724cd
commit 24353e6
Show file tree

Hide file tree

Showing 5 changed files with 13 additions and 23 deletions.
diff --git a/src/so_vits_svc_fork/__main__.py b/src/so_vits_svc_fork/__main__.py
@@ -281,7 +281,7 @@ def infer(
     "-fm",
     "--f0-method",
     type=click.Choice(["crepe", "crepe-tiny", "parselmouth", "dio", "harvest"]),
-    default="crepe",
+    default="dio",
     help="f0 prediction method",
 )
 @click.option("-p", "--pad-seconds", type=float, default=0.02, help="pad seconds")
@@ -505,7 +505,7 @@ def pre_config(
     "-fm",
     "--f0-method",
     type=click.Choice(["crepe", "crepe-tiny", "parselmouth", "dio", "harvest"]),
-    default="crepe",
+    default="dio",
 )
 def pre_hubert(
     input_dir: Path,

diff --git a/src/so_vits_svc_fork/inference/infer_tool.py b/src/so_vits_svc_fork/inference/infer_tool.py
@@ -130,7 +130,7 @@ def get_unit_f0(
         speaker: int | str,
         f0_method: Literal[
             "crepe", "crepe-tiny", "parselmouth", "dio", "harvest"
-        ] = "crepe",
+        ] = "dio",
     ):
         f0 = utils.compute_f0(
             audio,
@@ -172,7 +172,7 @@ def infer(
         noise_scale: float = 0.4,
         f0_method: Literal[
             "crepe", "crepe-tiny", "parselmouth", "dio", "harvest"
-        ] = "crepe",
+        ] = "dio",
     ) -> tuple[torch.Tensor, int]:
         audio = audio.astype(np.float32)
         # get speaker id
@@ -240,7 +240,7 @@ def infer_silence(
         noise_scale: float = 0.4,
         f0_method: Literal[
             "crepe", "crepe-tiny", "parselmouth", "dio", "harvest"
-        ] = "crepe",
+        ] = "dio",
         # slice config
         db_thresh: int = -40,
         pad_seconds: float = 0.5,
@@ -459,7 +459,7 @@ def infer(
         noise_scale: float = 0.4,
         f0_method: Literal[
             "crepe", "crepe-tiny", "parselmouth", "dio", "harvest"
-        ] = "crepe",
+        ] = "dio",
         # slice config
         db_thresh: int = -40,
         pad_seconds: float = 0.5,
@@ -519,7 +519,7 @@ def process(
         noise_scale: float = 0.4,
         f0_method: Literal[
             "crepe", "crepe-tiny", "parselmouth", "dio", "harvest"
-        ] = "crepe",
+        ] = "dio",
         # slice config
         db_thresh: int = -40,
         chunk_seconds: float = 0.5,

diff --git a/src/so_vits_svc_fork/inference_main.py b/src/so_vits_svc_fork/inference_main.py
@@ -29,9 +29,7 @@ def infer(
     auto_predict_f0: bool = False,
     cluster_infer_ratio: float = 0,
     noise_scale: float = 0.4,
-    f0_method: Literal[
-        "crepe", "crepe-tiny", "parselmouth", "dio", "harvest"
-    ] = "crepe",
+    f0_method: Literal["crepe", "crepe-tiny", "parselmouth", "dio", "harvest"] = "dio",
     # slice config
     db_thresh: int = -40,
     pad_seconds: float = 0.5,
@@ -83,9 +81,7 @@ def realtime(
     auto_predict_f0: bool = False,
     cluster_infer_ratio: float = 0,
     noise_scale: float = 0.4,
-    f0_method: Literal[
-        "crepe", "crepe-tiny", "parselmouth", "dio", "harvest"
-    ] = "crepe",
+    f0_method: Literal["crepe", "crepe-tiny", "parselmouth", "dio", "harvest"] = "dio",
     # slice config
     db_thresh: int = -40,
     pad_seconds: float = 0.5,

diff --git a/src/so_vits_svc_fork/preprocess_hubert_f0.py b/src/so_vits_svc_fork/preprocess_hubert_f0.py
@@ -23,9 +23,7 @@ def _process_one(
     sampling_rate: int,
     hop_length: int,
     device: Literal["cuda", "cpu"] = "cuda",
-    f0_method: Literal[
-        "crepe", "crepe-tiny", "parselmouth", "dio", "harvest"
-    ] = "crepe",
+    f0_method: Literal["crepe", "crepe-tiny", "parselmouth", "dio", "harvest"] = "dio",
     force_rebuild: bool = False,
 ):
     wav, sr = librosa.load(filepath, sr=sampling_rate)
@@ -59,9 +57,7 @@ def _process_batch(
     sampling_rate: int,
     hop_length: int,
     pbar_position: int,
-    f0_method: Literal[
-        "crepe", "crepe-tiny", "parselmouth", "dio", "harvest"
-    ] = "crepe",
+    f0_method: Literal["crepe", "crepe-tiny", "parselmouth", "dio", "harvest"] = "dio",
     force_rebuild: bool = False,
 ):
     device = "cuda" if torch.cuda.is_available() else "cpu"
@@ -83,9 +79,7 @@ def preprocess_hubert_f0(
     input_dir: Path | str,
     config_path: Path | str,
     n_jobs: int = 4,
-    f0_method: Literal[
-        "crepe", "crepe-tiny", "parselmouth", "dio", "harvest"
-    ] = "crepe",
+    f0_method: Literal["crepe", "crepe-tiny", "parselmouth", "dio", "harvest"] = "dio",
     force_rebuild: bool = False,
 ):
     input_dir = Path(input_dir)

diff --git a/src/so_vits_svc_fork/utils.py b/src/so_vits_svc_fork/utils.py
@@ -243,7 +243,7 @@ def compute_f0(
     p_len: None | int = None,
     sampling_rate: int = 44100,
     hop_length: int = 512,
-    method: Literal["crepe", "crepe-tiny", "parselmouth", "dio", "harvest"] = "crepe",
+    method: Literal["crepe", "crepe-tiny", "parselmouth", "dio", "harvest"] = "dio",
     **kwargs,
 ):
     with timer() as t: