isort all imports

TTS/bin/compute_attention_masks.py

@@ -1,12 +1,13 @@
 import argparse
 import importlib
 import os
+from argparse import RawTextHelpFormatter
 
 import numpy as np
 import torch
 from torch.utils.data import DataLoader
 from tqdm import tqdm
-from argparse import RawTextHelpFormatter
+
 from TTS.tts.datasets.TTSDataset import MyDataset
 from TTS.tts.utils.generic_utils import setup_model
 from TTS.tts.utils.io import load_checkpoint
@@ -14,7 +15,6 @@ from TTS.tts.utils.text.symbols import make_symbols, phonemes, symbols
 from TTS.utils.audio import AudioProcessor
 from TTS.utils.io import load_config
 
-
 if __name__ == "__main__":
     # pylint: disable=bad-option-value
     parser = argparse.ArgumentParser(

TTS/bin/compute_embeddings.py

@@ -3,14 +3,14 @@ import glob
 import os
 
 import numpy as np
+import torch
 from tqdm import tqdm
 
-import torch
 from TTS.speaker_encoder.model import SpeakerEncoder
+from TTS.tts.datasets.preprocess import load_meta_data
+from TTS.tts.utils.speakers import save_speaker_mapping
 from TTS.utils.audio import AudioProcessor
 from TTS.utils.io import load_config
-from TTS.tts.utils.speakers import save_speaker_mapping
-from TTS.tts.datasets.preprocess import load_meta_data
 
 parser = argparse.ArgumentParser(
     description='Compute embedding vectors for each wav file in a dataset. If "target_dataset" is defined, it generates "speakers.json" necessary for training a multi-speaker model.'

TTS/bin/compute_statistics.py

@@ -1,16 +1,16 @@
 #!/usr/bin/env python3
 # -*- coding: utf-8 -*-
 
-import os
-import glob
 import argparse
+import glob
+import os
 
 import numpy as np
 from tqdm import tqdm
 
 from TTS.tts.datasets.preprocess import load_meta_data
-from TTS.utils.io import load_config
 from TTS.utils.audio import AudioProcessor
+from TTS.utils.io import load_config
 
 
 def main():

TTS/bin/convert_melgan_tflite.py

@@ -7,7 +7,6 @@ from TTS.vocoder.tf.utils.generic_utils import setup_generator
 from TTS.vocoder.tf.utils.io import load_checkpoint
 from TTS.vocoder.tf.utils.tflite import convert_melgan_to_tflite
 
-
 parser = argparse.ArgumentParser()
 parser.add_argument("--tf_model", type=str, help="Path to target torch model to be converted to TF.")
 parser.add_argument("--config_path", type=str, help="Path to config file of torch model.")

TTS/bin/convert_melgan_torch_to_tf.py

@@ -1,17 +1,14 @@
 import argparse
-from difflib import SequenceMatcher
 import os
+from difflib import SequenceMatcher
 
 import numpy as np
 import tensorflow as tf
 import torch
 
 from TTS.utils.io import load_config
-from TTS.vocoder.tf.utils.convert_torch_to_tf_utils import (
-    compare_torch_tf,
-    convert_tf_name,
-    transfer_weights_torch_to_tf,
-)
+from TTS.vocoder.tf.utils.convert_torch_to_tf_utils import (compare_torch_tf, convert_tf_name,
+                                                            transfer_weights_torch_to_tf)
 from TTS.vocoder.tf.utils.generic_utils import setup_generator as setup_tf_generator
 from TTS.vocoder.tf.utils.io import save_checkpoint
 from TTS.vocoder.utils.generic_utils import setup_generator

TTS/bin/convert_tacotron2_tflite.py

@@ -2,12 +2,11 @@
 
 import argparse
 
-from TTS.utils.io import load_config
-from TTS.tts.utils.text.symbols import symbols, phonemes
 from TTS.tts.tf.utils.generic_utils import setup_model
 from TTS.tts.tf.utils.io import load_checkpoint
 from TTS.tts.tf.utils.tflite import convert_tacotron2_to_tflite
-
+from TTS.tts.utils.text.symbols import phonemes, symbols
+from TTS.utils.io import load_config
 
 parser = argparse.ArgumentParser()
 parser.add_argument("--tf_model", type=str, help="Path to target torch model to be converted to TF.")

TTS/bin/convert_tacotron2_torch_to_tf.py

@@ -1,12 +1,13 @@
 import argparse
-from difflib import SequenceMatcher
 import os
 import sys
+from difflib import SequenceMatcher
 from pprint import pprint
 
 import numpy as np
 import tensorflow as tf
 import torch
+
 from TTS.tts.tf.models.tacotron2 import Tacotron2
 from TTS.tts.tf.utils.convert_torch_to_tf_utils import compare_torch_tf, convert_tf_name, transfer_weights_torch_to_tf
 from TTS.tts.tf.utils.generic_utils import save_checkpoint

TTS/bin/distribute.py

@@ -1,12 +1,13 @@
 #!/usr/bin/env python3
 # -*- coding: utf-8 -*-
 
-import os
-import sys
-import pathlib
-import time
-import subprocess
 import argparse
+import os
+import pathlib
+import subprocess
+import sys
+import time
+
 import torch
 
 

TTS/bin/find_unique_chars.py

@@ -1,6 +1,6 @@
 """Find all the unique characters in a dataset"""
-import os
 import argparse
+import os
 from argparse import RawTextHelpFormatter
 
 from TTS.tts.datasets.preprocess import get_preprocessor_by_name

TTS/bin/resample.py

@@ -1,10 +1,11 @@
 import argparse
 import glob
 import os
-import librosa
-from distutils.dir_util import copy_tree
 from argparse import RawTextHelpFormatter
+from distutils.dir_util import copy_tree
 from multiprocessing import Pool
+
+import librosa
 from tqdm import tqdm
 
 

TTS/bin/synthesize.py

@@ -4,7 +4,6 @@
 import argparse
 import sys
 from argparse import RawTextHelpFormatter
-
 # pylint: disable=redefined-outer-name, unused-argument
 from pathlib import Path
 

TTS/bin/train_align_tts.py

@@ -12,6 +12,7 @@ import torch
 from torch.nn.parallel import DistributedDataParallel as DDP_th
 from torch.utils.data import DataLoader
 from torch.utils.data.distributed import DistributedSampler
+
 from TTS.tts.datasets.preprocess import load_meta_data
 from TTS.tts.datasets.TTSDataset import MyDataset
 from TTS.tts.layers.losses import AlignTTSLoss

TTS/bin/train_encoder.py

@@ -9,6 +9,7 @@ import traceback
 
 import torch
 from torch.utils.data import DataLoader
+
 from TTS.speaker_encoder.dataset import MyDataset
 from TTS.speaker_encoder.losses import AngleProtoLoss, GE2ELoss
 from TTS.speaker_encoder.model import SpeakerEncoder
@@ -16,13 +17,8 @@ from TTS.speaker_encoder.utils.generic_utils import check_config_speaker_encoder
 from TTS.speaker_encoder.utils.visual import plot_embeddings
 from TTS.tts.datasets.preprocess import load_meta_data
 from TTS.utils.audio import AudioProcessor
-from TTS.utils.generic_utils import (
-    count_parameters,
-    create_experiment_folder,
-    get_git_branch,
-    remove_experiment_folder,
-    set_init_dict,
-)
+from TTS.utils.generic_utils import (count_parameters, create_experiment_folder, get_git_branch,
+                                     remove_experiment_folder, set_init_dict)
 from TTS.utils.io import copy_model_files, load_config
 from TTS.utils.radam import RAdam
 from TTS.utils.tensorboard_logger import TensorboardLogger

TTS/bin/train_glow_tts.py

@@ -8,13 +8,11 @@ import traceback
 from random import randrange
 
 import torch
-
 # DISTRIBUTED
 from torch.nn.parallel import DistributedDataParallel as DDP_th
 from torch.utils.data import DataLoader
 from torch.utils.data.distributed import DistributedSampler
 
-from TTS.utils.arguments import parse_arguments, process_args
 from TTS.tts.datasets.preprocess import load_meta_data
 from TTS.tts.datasets.TTSDataset import MyDataset
 from TTS.tts.layers.losses import GlowTTSLoss
@@ -25,6 +23,7 @@ from TTS.tts.utils.speakers import parse_speakers
 from TTS.tts.utils.synthesis import synthesis
 from TTS.tts.utils.text.symbols import make_symbols, phonemes, symbols
 from TTS.tts.utils.visual import plot_alignment, plot_spectrogram
+from TTS.utils.arguments import parse_arguments, process_args
 from TTS.utils.audio import AudioProcessor
 from TTS.utils.distribute import init_distributed, reduce_tensor
 from TTS.utils.generic_utils import KeepAverage, count_parameters, remove_experiment_folder, set_init_dict

TTS/bin/train_speedy_speech.py

@@ -5,16 +5,15 @@ import os
 import sys
 import time
 import traceback
-import numpy as np
 from random import randrange
 
+import numpy as np
 import torch
-from TTS.utils.arguments import parse_arguments, process_args
-
 # DISTRIBUTED
 from torch.nn.parallel import DistributedDataParallel as DDP_th
 from torch.utils.data import DataLoader
 from torch.utils.data.distributed import DistributedSampler
+
 from TTS.tts.datasets.preprocess import load_meta_data
 from TTS.tts.datasets.TTSDataset import MyDataset
 from TTS.tts.layers.losses import SpeedySpeechLoss
@@ -25,6 +24,7 @@ from TTS.tts.utils.speakers import parse_speakers
 from TTS.tts.utils.synthesis import synthesis
 from TTS.tts.utils.text.symbols import make_symbols, phonemes, symbols
 from TTS.tts.utils.visual import plot_alignment, plot_spectrogram
+from TTS.utils.arguments import parse_arguments, process_args
 from TTS.utils.audio import AudioProcessor
 from TTS.utils.distribute import init_distributed, reduce_tensor
 from TTS.utils.generic_utils import KeepAverage, count_parameters, remove_experiment_folder, set_init_dict

TTS/bin/train_tacotron.py

@@ -10,7 +10,7 @@ from random import randrange
 import numpy as np
 import torch
 from torch.utils.data import DataLoader
-from TTS.utils.arguments import parse_arguments, process_args
+
 from TTS.tts.datasets.preprocess import load_meta_data
 from TTS.tts.datasets.TTSDataset import MyDataset
 from TTS.tts.layers.losses import TacotronLoss
@@ -21,18 +21,13 @@ from TTS.tts.utils.speakers import parse_speakers
 from TTS.tts.utils.synthesis import synthesis
 from TTS.tts.utils.text.symbols import make_symbols, phonemes, symbols
 from TTS.tts.utils.visual import plot_alignment, plot_spectrogram
+from TTS.utils.arguments import parse_arguments, process_args
 from TTS.utils.audio import AudioProcessor
 from TTS.utils.distribute import DistributedSampler, apply_gradient_allreduce, init_distributed, reduce_tensor
 from TTS.utils.generic_utils import KeepAverage, count_parameters, remove_experiment_folder, set_init_dict
 from TTS.utils.radam import RAdam
-from TTS.utils.training import (
-    NoamLR,
-    adam_weight_decay,
-    check_update,
-    gradual_training_scheduler,
-    set_weight_decay,
-    setup_torch_training_env,
-)
+from TTS.utils.training import (NoamLR, adam_weight_decay, check_update, gradual_training_scheduler, set_weight_decay,
+                                setup_torch_training_env)
 
 use_cuda, num_gpus = setup_torch_training_env(True, False)
 

TTS/bin/train_vocoder_gan.py

@@ -9,13 +9,16 @@ import traceback
 from inspect import signature
 
 import torch
+# DISTRIBUTED
+from torch.nn.parallel import DistributedDataParallel as DDP_th
 from torch.utils.data import DataLoader
+from torch.utils.data.distributed import DistributedSampler
+
 from TTS.utils.arguments import parse_arguments, process_args
 from TTS.utils.audio import AudioProcessor
+from TTS.utils.distribute import init_distributed
 from TTS.utils.generic_utils import KeepAverage, count_parameters, remove_experiment_folder, set_init_dict
-
 from TTS.utils.radam import RAdam
-
 from TTS.utils.training import setup_torch_training_env
 from TTS.vocoder.datasets.gan_dataset import GANDataset
 from TTS.vocoder.datasets.preprocess import load_wav_data, load_wav_feat_data
@@ -23,11 +26,6 @@ from TTS.vocoder.layers.losses import DiscriminatorLoss, GeneratorLoss
 from TTS.vocoder.utils.generic_utils import plot_results, setup_discriminator, setup_generator
 from TTS.vocoder.utils.io import save_best_model, save_checkpoint
 
-# DISTRIBUTED
-from torch.nn.parallel import DistributedDataParallel as DDP_th
-from torch.utils.data.distributed import DistributedSampler
-from TTS.utils.distribute import init_distributed
-
 use_cuda, num_gpus = setup_torch_training_env(True, True)
 
 

TTS/bin/train_vocoder_wavegrad.py

@@ -5,15 +5,15 @@ import os
 import sys
 import time
 import traceback
+
 import numpy as np
-
 import torch
-
 # DISTRIBUTED
 from torch.nn.parallel import DistributedDataParallel as DDP_th
 from torch.optim import Adam
 from torch.utils.data import DataLoader
 from torch.utils.data.distributed import DistributedSampler
+
 from TTS.utils.arguments import parse_arguments, process_args
 from TTS.utils.audio import AudioProcessor
 from TTS.utils.distribute import init_distributed

TTS/bin/train_vocoder_wavernn.py

@@ -2,33 +2,29 @@
 """Train WaveRNN vocoder model."""
 
 import os
-import sys
-import traceback
-import time
 import random
+import sys
+import time
+import traceback
 
 import torch
 from torch.utils.data import DataLoader
 
-# from torch.utils.data.distributed import DistributedSampler
-
-from TTS.utils.arguments import parse_arguments, process_args
 from TTS.tts.utils.visual import plot_spectrogram
+from TTS.utils.arguments import parse_arguments, process_args
 from TTS.utils.audio import AudioProcessor
+from TTS.utils.generic_utils import KeepAverage, count_parameters, remove_experiment_folder, set_init_dict
 from TTS.utils.radam import RAdam
 from TTS.utils.training import setup_torch_training_env
-from TTS.utils.generic_utils import (
-    KeepAverage,
-    count_parameters,
-    remove_experiment_folder,
-    set_init_dict,
-)
-from TTS.vocoder.datasets.wavernn_dataset import WaveRNNDataset
 from TTS.vocoder.datasets.preprocess import load_wav_data, load_wav_feat_data
+from TTS.vocoder.datasets.wavernn_dataset import WaveRNNDataset
 from TTS.vocoder.utils.distribution import discretized_mix_logistic_loss, gaussian_loss
 from TTS.vocoder.utils.generic_utils import setup_generator
 from TTS.vocoder.utils.io import save_best_model, save_checkpoint
 
+# from torch.utils.data.distributed import DistributedSampler
+
+
 
 use_cuda, num_gpus = setup_torch_training_env(True, True)
 

TTS/bin/tune_wavegrad.py

@@ -6,6 +6,7 @@ import numpy as np
 import torch
 from torch.utils.data import DataLoader
 from tqdm import tqdm
+
 from TTS.utils.audio import AudioProcessor
 from TTS.utils.io import load_config
 from TTS.vocoder.datasets.preprocess import load_wav_data

TTS/server/server.py

@@ -1,14 +1,15 @@
 #!flask/bin/python
 import argparse
+import io
 import os
 import sys
-import io
 from pathlib import Path
 
 from flask import Flask, render_template, request, send_file
-from TTS.utils.synthesizer import Synthesizer
-from TTS.utils.manage import ModelManager
+
 from TTS.utils.io import load_config
+from TTS.utils.manage import ModelManager
+from TTS.utils.synthesizer import Synthesizer
 
 
 def create_argparser():

TTS/speaker_encoder/utils/generic_utils.py

@@ -3,6 +3,7 @@ import os
 import re
 
 import torch
+
 from TTS.speaker_encoder.model import SpeakerEncoder
 from TTS.utils.generic_utils import check_argument
 

TTS/speaker_encoder/utils/prepare_voxceleb.py

@@ -19,15 +19,17 @@
 # pylint: disable=too-many-locals, too-many-statements, too-many-arguments, too-many-instance-attributes
 """ voxceleb 1 & 2 """
 
+import hashlib
 import os
+import subprocess
 import sys
 import zipfile
-import subprocess
-import hashlib
-import pandas
-from absl import logging
-import tensorflow as tf
+
 import soundfile as sf
+import tensorflow as tf
+from absl import logging
+
+import pandas
 
 gfile = tf.compat.v1.gfile
 

TTS/speaker_encoder/utils/visual.py

@@ -1,7 +1,7 @@
-import umap
-import numpy as np
 import matplotlib
 import matplotlib.pyplot as plt
+import numpy as np
+import umap
 
 matplotlib.use("Agg")
 

TTS/tts/datasets/TTSDataset.py

@@ -7,6 +7,7 @@ import numpy as np
 import torch
 import tqdm
 from torch.utils.data import Dataset
+
 from TTS.tts.utils.data import prepare_data, prepare_stop_target, prepare_tensor
 from TTS.tts.utils.text import pad_with_eos_bos, phoneme_to_sequence, text_to_sequence
 

TTS/tts/datasets/preprocess.py

@@ -7,6 +7,7 @@ from pathlib import Path
 from typing import List
 
 from tqdm import tqdm
+
 from TTS.tts.utils.generic_utils import split_dataset
 
 ####################

TTS/tts/layers/align_tts/duration_predictor.py

@@ -1,6 +1,7 @@
 from torch import nn
-from TTS.tts.layers.generic.transformer import FFTransformerBlock
+
 from TTS.tts.layers.generic.pos_encoding import PositionalEncoding
+from TTS.tts.layers.generic.transformer import FFTransformerBlock
 
 
 class DurationPredictor(nn.Module):

TTS/tts/layers/feed_forward/decoder.py

@@ -1,9 +1,10 @@
 import torch
 from torch import nn
-from TTS.tts.layers.generic.res_conv_bn import Conv1dBNBlock, ResidualConv1dBNBlock, Conv1dBN
+
+from TTS.tts.layers.generic.res_conv_bn import Conv1dBN, Conv1dBNBlock, ResidualConv1dBNBlock
+from TTS.tts.layers.generic.transformer import FFTransformerBlock
 from TTS.tts.layers.generic.wavenet import WNBlocks
 from TTS.tts.layers.glow_tts.transformer import RelativePositionTransformer
-from TTS.tts.layers.generic.transformer import FFTransformerBlock
 
 
 class WaveNetDecoder(nn.Module):

TTS/tts/layers/feed_forward/encoder.py

@@ -1,8 +1,8 @@
 from torch import nn
 
-from TTS.tts.layers.glow_tts.transformer import RelativePositionTransformer
 from TTS.tts.layers.generic.res_conv_bn import ResidualConv1dBNBlock
 from TTS.tts.layers.generic.transformer import FFTransformerBlock
+from TTS.tts.layers.glow_tts.transformer import RelativePositionTransformer
 
 
 class RelativePositionTransformerEncoder(nn.Module):

TTS/tts/layers/generic/pos_encoding.py

@@ -1,6 +1,6 @@
-import torch
 import math
 
+import torch
 from torch import nn
 
 

TTS/tts/layers/glow_tts/decoder.py

@@ -1,8 +1,8 @@
 import torch
 from torch import nn
 
-from TTS.tts.layers.glow_tts.glow import InvConvNear, CouplingBlock
 from TTS.tts.layers.generic.normalization import ActNorm
+from TTS.tts.layers.glow_tts.glow import CouplingBlock, InvConvNear
 
 
 def squeeze(x, x_mask=None, num_sqz=2):

TTS/tts/layers/glow_tts/encoder.py

@@ -1,14 +1,15 @@
 import math
+
 import torch
 from torch import nn
 
-from TTS.tts.layers.glow_tts.transformer import RelativePositionTransformer
 from TTS.tts.layers.generic.gated_conv import GatedConvBlock
-from TTS.tts.utils.generic_utils import sequence_mask
-from TTS.tts.layers.glow_tts.glow import ResidualConv1dLayerNormBlock
-from TTS.tts.layers.glow_tts.duration_predictor import DurationPredictor
-from TTS.tts.layers.generic.time_depth_sep_conv import TimeDepthSeparableConvBlock
 from TTS.tts.layers.generic.res_conv_bn import ResidualConv1dBNBlock
+from TTS.tts.layers.generic.time_depth_sep_conv import TimeDepthSeparableConvBlock
+from TTS.tts.layers.glow_tts.duration_predictor import DurationPredictor
+from TTS.tts.layers.glow_tts.glow import ResidualConv1dLayerNormBlock
+from TTS.tts.layers.glow_tts.transformer import RelativePositionTransformer
+from TTS.tts.utils.generic_utils import sequence_mask
 
 
 class Encoder(nn.Module):

TTS/tts/layers/glow_tts/glow.py

@@ -1,6 +1,7 @@
 import torch
 from torch import nn
 from torch.nn import functional as F
+
 from TTS.tts.layers.generic.wavenet import WN
 
 from ..generic.normalization import LayerNorm

TTS/tts/layers/glow_tts/monotonic_align/__init__.py

@@ -1,6 +1,7 @@
 import numpy as np
 import torch
 from torch.nn import functional as F
+
 from TTS.tts.utils.generic_utils import sequence_mask
 
 try:

TTS/tts/layers/glow_tts/transformer.py

@@ -1,4 +1,5 @@
 import math
+
 import torch
 from torch import nn
 from torch.nn import functional as F

TTS/tts/layers/losses.py

@@ -1,8 +1,10 @@
 import math
+
 import numpy as np
 import torch
 from torch import nn
 from torch.nn import functional
+
 from TTS.tts.utils.generic_utils import sequence_mask
 from TTS.tts.utils.ssim import ssim
 

TTS/tts/layers/tacotron/attentions.py

@@ -1,9 +1,9 @@
 import torch
+from scipy.stats import betabinom
 from torch import nn
 from torch.nn import functional as F
 
 from TTS.tts.layers.tacotron.common_layers import Linear
-from scipy.stats import betabinom
 
 
 class LocationLayer(nn.Module):

TTS/tts/layers/tacotron/tacotron.py

@@ -1,8 +1,9 @@
 # coding: utf-8
 import torch
 from torch import nn
-from .common_layers import Prenet
+
 from .attentions import init_attn
+from .common_layers import Prenet
 
 
 class BatchNormConv1d(nn.Module):

TTS/tts/layers/tacotron/tacotron2.py

@@ -1,8 +1,10 @@
 import torch
 from torch import nn
 from torch.nn import functional as F
-from .common_layers import Prenet, Linear
+
 from .attentions import init_attn
+from .common_layers import Linear, Prenet
+
 
 # NOTE: linter has a problem with the current TF release
 # pylint: disable=no-value-for-parameter

TTS/tts/models/align_tts.py

@@ -1,12 +1,13 @@
 import torch
 import torch.nn as nn
-from TTS.tts.layers.generic.pos_encoding import PositionalEncoding
+
+from TTS.tts.layers.align_tts.mdn import MDNBlock
+from TTS.tts.layers.feed_forward.decoder import Decoder
 from TTS.tts.layers.feed_forward.duration_predictor import DurationPredictor
+from TTS.tts.layers.feed_forward.encoder import Encoder
+from TTS.tts.layers.generic.pos_encoding import PositionalEncoding
 from TTS.tts.layers.glow_tts.monotonic_align import generate_path, maximum_path
 from TTS.tts.utils.generic_utils import sequence_mask
-from TTS.tts.layers.align_tts.mdn import MDNBlock
-from TTS.tts.layers.feed_forward.encoder import Encoder
-from TTS.tts.layers.feed_forward.decoder import Decoder
 
 
 class AlignTTS(nn.Module):

TTS/tts/models/glow_tts.py

@@ -1,12 +1,13 @@
 import math
+
 import torch
 from torch import nn
 from torch.nn import functional as F
 
-from TTS.tts.layers.glow_tts.encoder import Encoder
 from TTS.tts.layers.glow_tts.decoder import Decoder
+from TTS.tts.layers.glow_tts.encoder import Encoder
+from TTS.tts.layers.glow_tts.monotonic_align import generate_path, maximum_path
 from TTS.tts.utils.generic_utils import sequence_mask
-from TTS.tts.layers.glow_tts.monotonic_align import maximum_path, generate_path
 
 
 class GlowTTS(nn.Module):

TTS/tts/models/speedy_speech.py

@@ -1,11 +1,12 @@
 import torch
 from torch import nn
+
 from TTS.tts.layers.feed_forward.decoder import Decoder
 from TTS.tts.layers.feed_forward.duration_predictor import DurationPredictor
 from TTS.tts.layers.feed_forward.encoder import Encoder
 from TTS.tts.layers.generic.pos_encoding import PositionalEncoding
-from TTS.tts.utils.generic_utils import sequence_mask
 from TTS.tts.layers.glow_tts.monotonic_align import generate_path
+from TTS.tts.utils.generic_utils import sequence_mask
 
 
 class SpeedySpeech(nn.Module):

TTS/tts/models/tacotron2.py

@@ -5,6 +5,7 @@ from TTS.tts.layers.tacotron.gst_layers import GST
 from TTS.tts.layers.tacotron.tacotron2 import Decoder, Encoder, Postnet
 from TTS.tts.models.tacotron_abstract import TacotronAbstract
 
+
 # TODO: match function arguments with tacotron
 class Tacotron2(TacotronAbstract):
     """Tacotron2 as in https://arxiv.org/abs/1712.05884

TTS/tts/tf/layers/tacotron/tacotron2.py

@@ -1,7 +1,8 @@
 import tensorflow as tf
 from tensorflow import keras
+
+from TTS.tts.tf.layers.tacotron.common_layers import Attention, Prenet
 from TTS.tts.tf.utils.tf_utils import shape_list
-from TTS.tts.tf.layers.tacotron.common_layers import Prenet, Attention
 
 
 # NOTE: linter has a problem with the current TF release

TTS/tts/tf/models/tacotron2.py

@@ -1,7 +1,7 @@
 import tensorflow as tf
 from tensorflow import keras
 
-from TTS.tts.tf.layers.tacotron.tacotron2 import Encoder, Decoder, Postnet
+from TTS.tts.tf.layers.tacotron.tacotron2 import Decoder, Encoder, Postnet
 from TTS.tts.tf.utils.tf_utils import shape_list
 
 

TTS/tts/tf/utils/generic_utils.py

@@ -1,6 +1,7 @@
 import datetime
 import importlib
 import pickle
+
 import numpy as np
 import tensorflow as tf
 

TTS/tts/tf/utils/io.py

@@ -1,5 +1,6 @@
-import pickle
 import datetime
+import pickle
+
 import tensorflow as tf
 
 

TTS/tts/utils/chinese_mandarin/numbers.py

@@ -5,8 +5,8 @@
 # This uses Python 3, but it's easy to port to Python 2 by changing
 # strings to u'xx'.
 
-import re
 import itertools
+import re
 
 
 def _num2chinese(num: str, big=False, simp=True, o=False, twoalt=False) -> str:

TTS/tts/utils/chinese_mandarin/phonemizer.py

@@ -1,13 +1,11 @@
 from typing import List
 
+import jieba
 import pypinyin
 
 from .pinyinToPhonemes import PINYIN_DICT
 
 
-import jieba
-
-
 def _chinese_character_to_pinyin(text: str) -> List[str]:
     pinyins = pypinyin.pinyin(text, style=pypinyin.Style.TONE3, heteronym=False, neutral_tone_with_five=True)
     pinyins_flat_list = [item for sublist in pinyins for item in sublist]

TTS/tts/utils/generic_utils.py

@@ -1,9 +1,10 @@
-import re
-import torch
 import importlib
-import numpy as np
+import re
 from collections import Counter
 
+import numpy as np
+import torch
+
 from TTS.utils.generic_utils import check_argument
 
 

TTS/tts/utils/io.py

@@ -1,8 +1,9 @@
-import os
-import torch
 import datetime
+import os
 import pickle as pickle_tts
 
+import torch
+
 from TTS.utils.io import RenamingUnpickler
 
 

TTS/tts/utils/speakers.py

@@ -1,5 +1,5 @@
-import os
 import json
+import os
 
 
 def make_speakers_json_path(out_path):

TTS/tts/utils/synthesis.py

@@ -1,14 +1,16 @@
 import os
 
-os.environ["TF_CPP_MIN_LOG_LEVEL"] = "3"
+import numpy as np
 import pkg_resources
+import torch
+
+from .text import phoneme_to_sequence, text_to_sequence
+
+os.environ["TF_CPP_MIN_LOG_LEVEL"] = "3"
 
 installed = {pkg.key for pkg in pkg_resources.working_set}  # pylint: disable=not-an-iterable
 if "tensorflow" in installed or "tensorflow-gpu" in installed:
     import tensorflow as tf
-import torch
-import numpy as np
-from .text import text_to_sequence, phoneme_to_sequence
 
 
 def text_to_seqvec(text, CONFIG):

TTS/tts/utils/text/__init__.py

@@ -5,10 +5,10 @@ import re
 import phonemizer
 from packaging import version
 from phonemizer.phonemize import phonemize
+
+from TTS.tts.utils.chinese_mandarin.phonemizer import chinese_text_to_phonemes
 from TTS.tts.utils.text import cleaners
 from TTS.tts.utils.text.symbols import _bos, _eos, _punctuations, make_symbols, phonemes, symbols
-from TTS.tts.utils.chinese_mandarin.phonemizer import chinese_text_to_phonemes
-
 
 # pylint: disable=unnecessary-comprehension
 # Mappings from symbol to numeric ID and vice versa:

TTS/tts/utils/text/cleaners.py

@@ -11,12 +11,14 @@ hyperparameter. Some cleaners are English-specific. You'll typically want to use
 """
 
 import re
+
 from unidecode import unidecode
-from .number_norm import normalize_numbers
-from .abbreviations import abbreviations_en, abbreviations_fr
-from .time import expand_time_english
+
 from TTS.tts.utils.chinese_mandarin.numbers import replace_numbers_to_characters_in_text
 
+from .abbreviations import abbreviations_en, abbreviations_fr
+from .number_norm import normalize_numbers
+from .time import expand_time_english
 
 # Regular expression matching whitespace:
 _whitespace_re = re.compile(r"\s+")

TTS/tts/utils/text/number_norm.py

@@ -1,9 +1,10 @@
 """ from https://github.com/keithito/tacotron """
 
-import inflect
 import re
 from typing import Dict
 
+import inflect
+
 _inflect = inflect.engine()
 _comma_number_re = re.compile(r"([0-9][0-9\,]+[0-9])")
 _decimal_number_re = re.compile(r"([0-9]+\.[0-9]+)")

TTS/tts/utils/text/time.py

@@ -1,4 +1,5 @@
 import re
+
 import inflect
 
 _inflect = inflect.engine()

TTS/tts/utils/visual.py

@@ -1,12 +1,13 @@
 import librosa
 import matplotlib
+import matplotlib.pyplot as plt
 import numpy as np
 import torch
 
-matplotlib.use("Agg")
-import matplotlib.pyplot as plt
 from TTS.tts.utils.text import phoneme_to_sequence, sequence_to_phoneme
 
+matplotlib.use("Agg")
+
 
 def plot_alignment(alignment, info=None, fig_size=(16, 10), title=None, output_fig=False):
     if isinstance(alignment, torch.Tensor):

TTS/utils/arguments.py

@@ -4,11 +4,12 @@
 
 import argparse
 import glob
+import json
 import os
 import re
-import json
 
 import torch
+
 from TTS.tts.utils.text.symbols import parse_symbols
 from TTS.utils.console_logger import ConsoleLogger
 from TTS.utils.generic_utils import create_experiment_folder, get_git_branch

TTS/utils/audio.py

@@ -1,12 +1,13 @@
 import librosa
-import soundfile as sf
 import numpy as np
 import scipy.io.wavfile
 import scipy.signal
+import soundfile as sf
+
+from TTS.tts.utils.data import StandardScaler
 
 # import pyworld as pw
 
-from TTS.tts.utils.data import StandardScaler
 
 # pylint: disable=too-many-public-methods
 class AudioProcessor(object):

TTS/utils/console_logger.py

@@ -1,6 +1,6 @@
 import datetime
-from TTS.utils.io import AttrDict
 
+from TTS.utils.io import AttrDict
 
 tcolors = AttrDict(
     {

TTS/utils/io.py

@@ -1,10 +1,11 @@
-import os
-import re
 import json
-import yaml
+import os
 import pickle as pickle_tts
+import re
 from shutil import copyfile
 
+import yaml
+
 
 class RenamingUnpickler(pickle_tts.Unpickler):
     """Overload default pickler to solve module renaming problem"""

TTS/utils/manage.py

@@ -7,6 +7,7 @@ from shutil import copyfile
 
 import gdown
 import requests
+
 from TTS.utils.generic_utils import get_user_data_dir
 from TTS.utils.io import load_config
 

TTS/utils/radam.py

@@ -1,6 +1,7 @@
 # from https://github.com/LiyuanLucasLiu/RAdam
 
 import math
+
 import torch
 from torch.optim.optimizer import Optimizer
 

TTS/utils/synthesizer.py

@@ -1,19 +1,18 @@
 import time
 
 import numpy as np
-import torch
 import pysbd
+import torch
 
-from TTS.utils.audio import AudioProcessor
-from TTS.utils.io import load_config
 from TTS.tts.utils.generic_utils import setup_model
 from TTS.tts.utils.speakers import load_speaker_mapping
-from TTS.vocoder.utils.generic_utils import setup_generator, interpolate_vocoder_input
-
 # pylint: disable=unused-wildcard-import
 # pylint: disable=wildcard-import
 from TTS.tts.utils.synthesis import synthesis, trim_silence
 from TTS.tts.utils.text import make_symbols, phonemes, symbols
+from TTS.utils.audio import AudioProcessor
+from TTS.utils.io import load_config
+from TTS.vocoder.utils.generic_utils import interpolate_vocoder_input, setup_generator
 
 
 class Synthesizer(object):

TTS/utils/tensorboard_logger.py

@@ -1,4 +1,5 @@
 import traceback
+
 from tensorboardX import SummaryWriter
 
 

TTS/utils/training.py

@@ -1,5 +1,5 @@
-import torch
 import numpy as np
+import torch
 
 
 def setup_torch_training_env(cudnn_enable, cudnn_benchmark):

TTS/vocoder/datasets/gan_dataset.py

@@ -1,11 +1,12 @@
-import os
 import glob
-import torch
+import os
 import random
-import numpy as np
-from torch.utils.data import Dataset
 from multiprocessing import Manager
 
+import numpy as np
+import torch
+from torch.utils.data import Dataset
+
 
 class GANDataset(Dataset):
     """

TTS/vocoder/datasets/preprocess.py

@@ -1,9 +1,9 @@
 import glob
 import os
 from pathlib import Path
-from tqdm import tqdm
 
 import numpy as np
+from tqdm import tqdm
 
 
 def preprocess_wav_files(out_path, config, ap):

TTS/vocoder/datasets/wavegrad_dataset.py

@@ -1,11 +1,12 @@
-import os
 import glob
-import torch
+import os
 import random
-import numpy as np
-from torch.utils.data import Dataset
 from multiprocessing import Manager
 
+import numpy as np
+import torch
+from torch.utils.data import Dataset
+
 
 class WaveGradDataset(Dataset):
     """

TTS/vocoder/datasets/wavernn_dataset.py

@@ -1,5 +1,5 @@
-import torch
 import numpy as np
+import torch
 from torch.utils.data import Dataset
 
 

TTS/vocoder/layers/losses.py

@@ -1,5 +1,5 @@
-import torch
 import librosa
+import torch
 from torch import nn
 from torch.nn import functional as F
 

TTS/vocoder/layers/pqmf.py

@@ -1,7 +1,6 @@
 import numpy as np
 import torch
 import torch.nn.functional as F
-
 from scipy import signal as sig
 
 

TTS/vocoder/models/multiband_melgan_generator.py

@@ -1,7 +1,7 @@
 import torch
 
-from TTS.vocoder.models.melgan_generator import MelganGenerator
 from TTS.vocoder.layers.pqmf import PQMF
+from TTS.vocoder.models.melgan_generator import MelganGenerator
 
 
 class MultibandMelganGenerator(MelganGenerator):

TTS/vocoder/models/parallel_wavegan_discriminator.py

@@ -1,4 +1,5 @@
 import math
+
 import torch
 from torch import nn
 

TTS/vocoder/models/parallel_wavegan_generator.py

@@ -1,4 +1,5 @@
 import math
+
 import numpy as np
 import torch
 

TTS/vocoder/models/wavegrad.py

@@ -3,7 +3,7 @@ import torch
 from torch import nn
 from torch.nn.utils import weight_norm
 
-from ..layers.wavegrad import DBlock, FiLM, UBlock, Conv1d
+from ..layers.wavegrad import Conv1d, DBlock, FiLM, UBlock
 
 
 class Wavegrad(nn.Module):

TTS/vocoder/models/wavernn.py

@@ -1,16 +1,14 @@
 import sys
+import time
+
+import numpy as np
 import torch
 import torch.nn as nn
-import numpy as np
 import torch.nn.functional as F
-import time
 
 # fix this
 from TTS.utils.audio import AudioProcessor as ap
-from TTS.vocoder.utils.distribution import (
-    sample_from_gaussian,
-    sample_from_discretized_mix_logistic,
-)
+from TTS.vocoder.utils.distribution import sample_from_discretized_mix_logistic, sample_from_gaussian
 
 
 def stream(string, variables):

TTS/vocoder/tf/layers/pqmf.py

@@ -1,6 +1,5 @@
 import numpy as np
 import tensorflow as tf
-
 from scipy import signal as sig
 
 

TTS/vocoder/tf/models/melgan_generator.py

@@ -1,11 +1,13 @@
 import logging
 import os
 
+import tensorflow as tf
+
+from TTS.vocoder.tf.layers.melgan import ReflectionPad1d, ResidualStack
+
 os.environ["TF_CPP_MIN_LOG_LEVEL"] = "3"  # FATAL
 logging.getLogger("tensorflow").setLevel(logging.FATAL)
 
-import tensorflow as tf
-from TTS.vocoder.tf.layers.melgan import ResidualStack, ReflectionPad1d
 
 
 # pylint: disable=too-many-ancestors

TTS/vocoder/tf/models/multiband_melgan_generator.py

@@ -1,7 +1,8 @@
 import tensorflow as tf
 
-from TTS.vocoder.tf.models.melgan_generator import MelganGenerator
 from TTS.vocoder.tf.layers.pqmf import PQMF
+from TTS.vocoder.tf.models.melgan_generator import MelganGenerator
+
 
 # pylint: disable=too-many-ancestors
 # pylint: disable=abstract-method

TTS/vocoder/tf/utils/generic_utils.py

@@ -1,5 +1,5 @@
-import re
 import importlib
+import re
 
 
 def to_camel(text):

TTS/vocoder/tf/utils/io.py

@@ -1,5 +1,6 @@
 import datetime
 import pickle
+
 import tensorflow as tf
 
 

TTS/vocoder/utils/distribution.py

@@ -1,8 +1,9 @@
-import numpy as np
 import math
+
+import numpy as np
 import torch
-from torch.distributions.normal import Normal
 import torch.nn.functional as F
+from torch.distributions.normal import Normal
 
 
 def gaussian_loss(y_hat, y, log_std_min=-7.0):

TTS/vocoder/utils/generic_utils.py

@@ -1,7 +1,8 @@
-import re
-import torch
 import importlib
+import re
+
 import numpy as np
+import torch
 from matplotlib import pyplot as plt
 
 from TTS.tts.utils.visual import plot_spectrogram

TTS/vocoder/utils/io.py

@@ -1,9 +1,10 @@
-import os
-import glob
-import torch
 import datetime
+import glob
+import os
 import pickle as pickle_tts
 
+import torch
+
 from TTS.utils.io import RenamingUnpickler
 
 

hubconf.py

@@ -1,8 +1,8 @@
 dependencies = ['torch', 'gdown', 'pysbd', 'phonemizer', 'unidecode', 'pypinyin']  # apt install espeak-ng
 import torch
 
-from TTS.utils.synthesizer import Synthesizer
 from TTS.utils.manage import ModelManager
+from TTS.utils.synthesizer import Synthesizer
 
 
 def tts(model_name='tts_models/en/ljspeech/tacotron2-DCA', vocoder_name=None, use_cuda=False):

notebooks/dataset_analysis/analyze.py

@@ -1,14 +1,16 @@
 # visualisation tools for mimic2
-import matplotlib.pyplot as plt
-from statistics import stdev, mode, mean, median
-from statistics import StatisticsError
 import argparse
-import os
 import csv
-import seaborn as sns
+import os
 import random
+from statistics import StatisticsError, mean, median, mode, stdev
+
+import matplotlib.pyplot as plt
+
+import seaborn as sns
 from text.cmudict import CMUDict
 
+
 def get_audio_seconds(frames):
     return (frames*12.5)/1000
 

setup.py

@@ -8,9 +8,8 @@ from distutils.version import LooseVersion
 import numpy
 import setuptools.command.build_py
 import setuptools.command.develop
-from setuptools import setup, Extension, find_packages
 from Cython.Build import cythonize
-
+from setuptools import Extension, find_packages, setup
 
 if LooseVersion(sys.version) < LooseVersion("3.6") or LooseVersion(sys.version) > LooseVersion("3.9"):
     raise RuntimeError(

tests/test_audio.py

@@ -2,7 +2,6 @@ import os
 import unittest
 
 from tests import get_tests_input_path, get_tests_output_path, get_tests_path
-
 from TTS.utils.audio import AudioProcessor
 from TTS.utils.io import load_config
 

tests/test_feed_forward_layers.py

@@ -1,4 +1,5 @@
 import torch
+
 from TTS.tts.layers.feed_forward.decoder import Decoder
 from TTS.tts.layers.feed_forward.encoder import Encoder
 from TTS.tts.utils.generic_utils import sequence_mask

tests/test_glow_tts.py

@@ -3,13 +3,13 @@ import os
 import unittest
 
 import torch
-from tests import get_tests_input_path
 from torch import optim
 
+from tests import get_tests_input_path
 from TTS.tts.layers.losses import GlowTTSLoss
 from TTS.tts.models.glow_tts import GlowTTS
-from TTS.utils.io import load_config
 from TTS.utils.audio import AudioProcessor
+from TTS.utils.io import load_config
 
 #pylint: disable=unused-variable
 

tests/test_layers.py

@@ -1,8 +1,9 @@
 import unittest
+
 import torch as T
 
-from TTS.tts.layers.tacotron.tacotron import Prenet, CBHG, Decoder, Encoder
 from TTS.tts.layers.losses import L1LossMasked, SSIMLoss
+from TTS.tts.layers.tacotron.tacotron import CBHG, Decoder, Encoder, Prenet
 from TTS.tts.utils.generic_utils import sequence_mask
 
 # pylint: disable=unused-variable

tests/test_loader.py

@@ -4,9 +4,9 @@ import unittest
 
 import numpy as np
 import torch
-from tests import get_tests_input_path, get_tests_output_path
 from torch.utils.data import DataLoader
 
+from tests import get_tests_input_path, get_tests_output_path
 from TTS.tts.datasets import TTSDataset
 from TTS.tts.datasets.preprocess import ljspeech
 from TTS.utils.audio import AudioProcessor

tests/test_preprocessors.py

@@ -1,7 +1,7 @@
-import unittest
 import os
-from tests import get_tests_input_path
+import unittest
 
+from tests import get_tests_input_path
 from TTS.tts.datasets.preprocess import common_voice
 
 

tests/test_speaker_encoder.py

@@ -2,9 +2,9 @@ import os
 import unittest
 
 import torch as T
-from tests import get_tests_input_path
 
-from TTS.speaker_encoder.losses import GE2ELoss, AngleProtoLoss
+from tests import get_tests_input_path
+from TTS.speaker_encoder.losses import AngleProtoLoss, GE2ELoss
 from TTS.speaker_encoder.model import SpeakerEncoder
 from TTS.utils.io import load_config
 

tests/test_speedy_speech_layers.py

@@ -1,8 +1,8 @@
 import torch
-from TTS.tts.layers.feed_forward.duration_predictor import DurationPredictor
-from TTS.tts.utils.generic_utils import sequence_mask
-from TTS.tts.models.speedy_speech import SpeedySpeech
 
+from TTS.tts.layers.feed_forward.duration_predictor import DurationPredictor
+from TTS.tts.models.speedy_speech import SpeedySpeech
+from TTS.tts.utils.generic_utils import sequence_mask
 
 use_cuda = torch.cuda.is_available()
 device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")

tests/test_symbols.py

@@ -2,6 +2,7 @@ import unittest
 
 from TTS.tts.utils.text import phonemes
 
+
 class SymbolsTest(unittest.TestCase):
     def test_uniqueness(self):  #pylint: disable=no-self-use
         assert sorted(phonemes) == sorted(list(set(phonemes))), " {} vs {} ".format(len(phonemes), len(set(phonemes)))

tests/test_synthesizer.py

@@ -2,11 +2,11 @@ import os
 import unittest
 
 from tests import get_tests_input_path, get_tests_output_path
-from TTS.utils.synthesizer import Synthesizer
 from TTS.tts.utils.generic_utils import setup_model
 from TTS.tts.utils.io import save_checkpoint
 from TTS.tts.utils.text.symbols import make_symbols, phonemes, symbols
 from TTS.utils.io import load_config
+from TTS.utils.synthesizer import Synthesizer
 
 
 class SynthesizerTest(unittest.TestCase):

tests/test_tacotron2_model.py

@@ -3,13 +3,13 @@ import os
 import unittest
 
 import torch
-from tests import get_tests_input_path
 from torch import nn, optim
 
+from tests import get_tests_input_path
 from TTS.tts.layers.losses import MSELossMasked
 from TTS.tts.models.tacotron2 import Tacotron2
-from TTS.utils.io import load_config
 from TTS.utils.audio import AudioProcessor
+from TTS.utils.io import load_config
 
 #pylint: disable=unused-variable