Fix lint: ruff E712

Author: Nugine

bins/svc/inference.py

@@ -50,7 +50,7 @@ def prepare_for_audio_file(args, cfg, num_workers=1):
     acoustic_extractor.extract_utt_acoustic_features_serial(
         metadata, temp_audio_dir, cfg
     )
-    if cfg.preprocess.use_min_max_norm_mel == True:
+    if cfg.preprocess.use_min_max_norm_mel is True:
         acoustic_extractor.cal_mel_min_max(
             dataset=audio_name, output_path=preprocess_path, cfg=cfg, metadata=metadata
         )

models/tta/ldm/inference_utils/vocoder.py

@@ -216,7 +216,7 @@ class DiscriminatorP(torch.nn.Module):
     def __init__(self, period, kernel_size=5, stride=3, use_spectral_norm=False):
         super(DiscriminatorP, self).__init__()
         self.period = period
-        norm_f = weight_norm if use_spectral_norm == False else spectral_norm
+        norm_f = weight_norm if use_spectral_norm is False else spectral_norm
         self.convs = nn.ModuleList(
             [
                 norm_f(
@@ -314,7 +314,7 @@ def forward(self, y, y_hat):
 class DiscriminatorS(torch.nn.Module):
     def __init__(self, use_spectral_norm=False):
         super(DiscriminatorS, self).__init__()
-        norm_f = weight_norm if use_spectral_norm == False else spectral_norm
+        norm_f = weight_norm if use_spectral_norm is False else spectral_norm
         self.convs = nn.ModuleList(
             [
                 norm_f(Conv1d(1, 128, 15, 1, padding=7)),

models/tts/naturalspeech2/ns2_dataset.py

@@ -36,7 +36,7 @@ def __init__(self, cfg, dataset, is_valid=False):
 
         self.cfg = cfg
 
-        assert cfg.preprocess.use_mel == False
+        assert cfg.preprocess.use_mel is False
         if cfg.preprocess.use_mel:
             self.utt2melspec_path = {}
             for utt_info in self.metadata:
@@ -52,7 +52,7 @@ def __init__(self, cfg, dataset, is_valid=False):
                     uid + ".npy",
                 )
 
-        assert cfg.preprocess.use_code == True
+        assert cfg.preprocess.use_code is True
         if cfg.preprocess.use_code:
             self.utt2code_path = {}
             for utt_info in self.metadata:
@@ -68,7 +68,7 @@ def __init__(self, cfg, dataset, is_valid=False):
                     uid + ".npy",
                 )
 
-        assert cfg.preprocess.use_spkid == True
+        assert cfg.preprocess.use_spkid is True
         if cfg.preprocess.use_spkid:
             self.utt2spkid = {}
             for utt_info in self.metadata:
@@ -78,7 +78,7 @@ def __init__(self, cfg, dataset, is_valid=False):
 
                 self.utt2spkid[utt] = utt_info["speaker"]
 
-        assert cfg.preprocess.use_pitch == True
+        assert cfg.preprocess.use_pitch is True
         if cfg.preprocess.use_pitch:
             self.utt2pitch_path = {}
             for utt_info in self.metadata:
@@ -94,7 +94,7 @@ def __init__(self, cfg, dataset, is_valid=False):
                     uid + ".npy",
                 )
 
-        assert cfg.preprocess.use_duration == True
+        assert cfg.preprocess.use_duration is True
         if cfg.preprocess.use_duration:
             self.utt2duration_path = {}
             for utt_info in self.metadata:
@@ -110,7 +110,7 @@ def __init__(self, cfg, dataset, is_valid=False):
                     uid + ".npy",
                 )
 
-        assert cfg.preprocess.use_phone == True
+        assert cfg.preprocess.use_phone is True
         if cfg.preprocess.use_phone:
             self.utt2phone = {}
             for utt_info in self.metadata:
@@ -120,7 +120,7 @@ def __init__(self, cfg, dataset, is_valid=False):
 
                 self.utt2phone[utt] = utt_info["phones"]
 
-        assert cfg.preprocess.use_len == True
+        assert cfg.preprocess.use_len is True
         if cfg.preprocess.use_len:
             self.utt2len = {}
             for utt_info in self.metadata:

models/tts/valle/valle_dataset.py

@@ -35,7 +35,7 @@ def __init__(self, cfg, dataset, is_valid=False):
 
         assert isinstance(dataset, str)
 
-        assert cfg.preprocess.use_acoustic_token == True
+        assert cfg.preprocess.use_acoustic_token is True
         if cfg.preprocess.use_acoustic_token:
             self.utt2acousticToken_path = {}
             for utt_info in self.metadata:
@@ -121,7 +121,7 @@ def __init__(self, args, cfg):
         super().__init__(args, cfg)
 
         # prepare data
-        assert cfg.preprocess.use_acoustic_token == True
+        assert cfg.preprocess.use_acoustic_token is True
         if cfg.preprocess.use_acoustic_token:
             self.utt2acousticToken = {}
             for utt_info in self.metadata:

models/vocoders/gan/discriminator/mpd.py

@@ -19,7 +19,7 @@ def __init__(self, cfg, period, kernel_size=5, stride=3, use_spectral_norm=False
         super(DiscriminatorP, self).__init__()
         self.period = period
         self.d_mult = cfg.model.mpd.discriminator_channel_mult_factor
-        norm_f = weight_norm if use_spectral_norm == False else spectral_norm
+        norm_f = weight_norm if use_spectral_norm is False else spectral_norm
         self.convs = nn.ModuleList(
             [
                 norm_f(
@@ -130,7 +130,7 @@ def __init__(self, period, kernel_size=5, stride=3, use_spectral_norm=False):
         super(DiscriminatorP_vits, self).__init__()
         self.period = period
         self.use_spectral_norm = use_spectral_norm
-        norm_f = weight_norm if use_spectral_norm == False else spectral_norm
+        norm_f = weight_norm if use_spectral_norm is False else spectral_norm
         self.convs = nn.ModuleList(
             [
                 norm_f(
@@ -207,7 +207,7 @@ def forward(self, x):
 class DiscriminatorS(torch.nn.Module):
     def __init__(self, use_spectral_norm=False):
         super(DiscriminatorS, self).__init__()
-        norm_f = weight_norm if use_spectral_norm == False else spectral_norm
+        norm_f = weight_norm if use_spectral_norm is False else spectral_norm
         self.convs = nn.ModuleList(
             [
                 norm_f(Conv1d(1, 16, 15, 1, padding=7)),

models/vocoders/gan/discriminator/mrd.py

@@ -27,7 +27,7 @@ def __init__(self, cfg, resolution):
         self.lrelu_slope = LRELU_SLOPE
 
         norm_f = (
-            weight_norm if cfg.model.mrd.use_spectral_norm == False else spectral_norm
+            weight_norm if cfg.model.mrd.use_spectral_norm is False else spectral_norm
         )
         if cfg.model.mrd.mrd_override:
             print(
@@ -37,7 +37,7 @@ def __init__(self, cfg, resolution):
             )
             norm_f = (
                 weight_norm
-                if cfg.model.mrd.mrd_use_spectral_norm == False
+                if cfg.model.mrd.mrd_use_spectral_norm is False
                 else spectral_norm
             )
         self.d_mult = cfg.model.mrd.discriminator_channel_mult_factor

models/vocoders/gan/discriminator/msd.py

@@ -19,7 +19,7 @@ class DiscriminatorS(nn.Module):
     def __init__(self, use_spectral_norm=False):
         super(DiscriminatorS, self).__init__()
 
-        norm_f = weight_norm if use_spectral_norm == False else spectral_norm
+        norm_f = weight_norm if use_spectral_norm is False else spectral_norm
 
         self.convs = nn.ModuleList(
             [

processors/data_augment.py

@@ -278,10 +278,10 @@ def wav_manipulation(
     ], "aug_type must be one of formant_shift, pitch_shift, time_stretch, equalizer"
 
     assert aug_type == "None" or (
-        formant_shift == False
-        and pitch_shift == False
-        and time_stretch == False
-        and equalizer == False
+        formant_shift is False
+        and pitch_shift is False
+        and time_stretch is False
+        and equalizer is False
     ), "if aug_type is specified, other argument must be False"
 
     if aug_type != "None":

utils/f0.py

@@ -91,7 +91,7 @@ def get_f0_features_using_pyin(audio, cfg):
         hop_length=cfg.hop_size,
     )
     # Set nan to 0
-    f0[voiced_flag == False] = 0
+    f0[voiced_flag is False] = 0
     return f0