Make style

TTS/tts/models/vits.py

@@ -204,6 +204,7 @@ def wav_to_mel(y, n_fft, num_mels, sample_rate, hop_length, win_length, fmin, fm
 # CONFIGS
 #############################
 
+
 class VitsAudioConfig(Coqpit):
     sample_rate: int = 22050
     win_length: int = 1024
@@ -212,6 +213,7 @@ class VitsAudioConfig(Coqpit):
     mel_fmin: int = 0
     mel_fmax: int = None
 
+
 ##############################
 # DATASET
 ##############################

recipes/ljspeech/vits_tts/train_vits.py

@@ -14,12 +14,7 @@ dataset_config = BaseDatasetConfig(
     name="ljspeech", meta_file_train="metadata.csv", path=os.path.join(output_path, "../LJSpeech-1.1/")
 )
 audio_config = VitsAudioConfig(
-    sample_rate=22050,
-    win_length=1024,
-    hop_length=256,
-    num_mels=80,
-    mel_fmin=0,
-    mel_fmax=None
+    sample_rate=22050, win_length=1024, hop_length=256, num_mels=80, mel_fmin=0, mel_fmax=None
 )
 
 config = VitsConfig(
@@ -43,7 +38,7 @@ config = VitsConfig(
     mixed_precision=True,
     output_path=output_path,
     datasets=[dataset_config],
-    cudnn_benchmark=False
+    cudnn_benchmark=False,
 )
 
 # INITIALIZE THE AUDIO PROCESSOR

recipes/vctk/vits/train_vits.py

@@ -17,12 +17,7 @@ dataset_config = BaseDatasetConfig(
 
 
 audio_config = VitsAudioConfig(
-    sample_rate=22050,
-    win_length=1024,
-    hop_length=256,
-    num_mels=80,
-    mel_fmin=0,
-    mel_fmax=None
+    sample_rate=22050, win_length=1024, hop_length=256, num_mels=80, mel_fmin=0, mel_fmax=None
 )
 
 vitsArgs = VitsArgs(
@@ -52,7 +47,7 @@ config = VitsConfig(
     max_text_len=325,  # change this if you have a larger VRAM than 16GB
     output_path=output_path,
     datasets=[dataset_config],
-    cudnn_benchmark=False
+    cudnn_benchmark=False,
 )
 
 # INITIALIZE THE AUDIO PROCESSOR