Lint fixs

TTS/bin/find_unique_phonemes.py

@@ -7,7 +7,7 @@ from tqdm.contrib.concurrent import process_map
 
 from TTS.config import load_config
 from TTS.tts.datasets import load_tts_samples
-from TTS.tts.utils.text import text2phone, phoneme_to_sequence
+from TTS.tts.utils.text import text2phone
 
 
 def compute_phonemes(item):
@@ -19,8 +19,8 @@ def compute_phonemes(item):
         return []
     return list(set(ph))
 
-
 def main():
+    # pylint: disable=W0601
     global c
     # pylint: disable=bad-option-value
     parser = argparse.ArgumentParser(

TTS/speaker_encoder/models/resnet.py

@@ -176,6 +176,7 @@ class ResNetSpeakerEncoder(nn.Module):
     def forward(self, x, l2_norm=False):
         with torch.no_grad():
             with torch.cuda.amp.autocast(enabled=False):
+                # if you torch spec compute it otherwise use the mel spec computed by the AP
                 if self.use_torch_spec:
                     x = self.torch_spec(x)
                 else:

TTS/tts/configs/vits_config.py

@@ -154,6 +154,6 @@ class VitsConfig(BaseTTSConfig):
     d_vector_dim: int = None
 
     def __post_init__(self):
-        for key in self.model_args.keys():
+        for key, val in self.model_args.items():
             if hasattr(self, key):
-                self[key] = self.model_args[key]
+                self[key] = val

TTS/utils/vad.py

@@ -139,4 +139,4 @@ def get_vad_speech_segments(audio, sample_rate, aggressiveness=2, padding_durati
     frames = list(frame_generator(30, audio, sample_rate))
     segments = vad_collector(sample_rate, 30, padding_duration_ms, vad, frames)
 
-    return segments
+    return segments