diff --git a/TTS/bin/train_encoder.py b/TTS/bin/train_encoder.py
index 8d377db241..914c729856 100644
--- a/TTS/bin/train_encoder.py
+++ b/TTS/bin/train_encoder.py
@@ -296,7 +296,7 @@ def main(args):  # pylint: disable=redefined-outer-name
         criterion, args.restore_step = model.load_checkpoint(
             c, args.restore_path, eval=False, use_cuda=use_cuda, criterion=criterion
         )
-        print(" > Model restored from step %d" % args.restore_step, flush=True)
+        print(f" > Model restored from step {args.restore_step}", flush=True)
     else:
         args.restore_step = 0
 
diff --git a/TTS/config/__init__.py b/TTS/config/__init__.py
index e47a141c73..401003504e 100644
--- a/TTS/config/__init__.py
+++ b/TTS/config/__init__.py
@@ -1,7 +1,7 @@
 import json
 import os
 import re
-from typing import Any, Dict, Union
+from typing import Any, Union
 
 import fsspec
 import yaml
@@ -58,7 +58,7 @@ def _process_model_name(config_dict: dict) -> str:
     """Format the model name as expected. It is a band-aid for the old `vocoder` model names.
 
     Args:
-        config_dict (Dict): A dictionary including the config fields.
+        config_dict (dict): A dictionary including the config fields.
 
     Returns:
         str: Formatted modelname.
diff --git a/TTS/encoder/models/base_encoder.py b/TTS/encoder/models/base_encoder.py
index 603481cc56..c6680c3a25 100644
--- a/TTS/encoder/models/base_encoder.py
+++ b/TTS/encoder/models/base_encoder.py
@@ -34,7 +34,7 @@ class BaseEncoder(nn.Module):
 
     # pylint: disable=W0102
     def __init__(self):
-        super(BaseEncoder, self).__init__()
+        super().__init__()
 
     def get_torch_mel_spectrogram_class(self, audio_config):
         return torch.nn.Sequential(
@@ -107,7 +107,7 @@ def get_criterion(self, c: Coqpit, num_classes=None):
         elif c.loss == "softmaxproto":
             criterion = SoftmaxAngleProtoLoss(c.model_params["proj_dim"], num_classes)
         else:
-            raise Exception("The %s  not is a loss supported" % c.loss)
+            raise Exception(f"The {c.loss}  not is a loss supported")
         return criterion
 
     def load_checkpoint(
diff --git a/TTS/encoder/models/resnet.py b/TTS/encoder/models/resnet.py
index 5eafcd6005..d7f3a2f4bd 100644
--- a/TTS/encoder/models/resnet.py
+++ b/TTS/encoder/models/resnet.py
@@ -7,7 +7,7 @@
 
 class SELayer(nn.Module):
     def __init__(self, channel, reduction=8):
-        super(SELayer, self).__init__()
+        super().__init__()
         self.avg_pool = nn.AdaptiveAvgPool2d(1)
         self.fc = nn.Sequential(
             nn.Linear(channel, channel // reduction),
@@ -27,7 +27,7 @@ class SEBasicBlock(nn.Module):
     expansion = 1
 
     def __init__(self, inplanes, planes, stride=1, downsample=None, reduction=8):
-        super(SEBasicBlock, self).__init__()
+        super().__init__()
         self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=3, stride=stride, padding=1, bias=False)
         self.bn1 = nn.BatchNorm2d(planes)
         self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, padding=1, bias=False)
@@ -73,7 +73,7 @@ def __init__(
         use_torch_spec=False,
         audio_config=None,
     ):
-        super(ResNetSpeakerEncoder, self).__init__()
+        super().__init__()
 
         self.encoder_type = encoder_type
         self.input_dim = input_dim
diff --git a/TTS/encoder/utils/prepare_voxceleb.py b/TTS/encoder/utils/prepare_voxceleb.py
index 18ca21c876..fe57874a99 100644
--- a/TTS/encoder/utils/prepare_voxceleb.py
+++ b/TTS/encoder/utils/prepare_voxceleb.py
@@ -82,17 +82,17 @@ def download_and_extract(directory, subset, urls):
                 continue
             logger.info("Downloading %s to %s", url, zip_filepath)
             subprocess.call(
-                "wget %s --user %s --password %s -O %s" % (url, USER["user"], USER["password"], zip_filepath),
+                "wget {} --user {} --password {} -O {}".format(url, USER["user"], USER["password"], zip_filepath),
                 shell=True,
             )
 
             statinfo = os.stat(zip_filepath)
-            logger.info("Successfully downloaded %s, size(bytes): %d" % (url, statinfo.st_size))
+            logger.info("Successfully downloaded %s, size(bytes): %d", url, statinfo.st_size)
 
         # concatenate all parts into zip files
         if ".zip" not in zip_filepath:
             zip_filepath = "_".join(zip_filepath.split("_")[:-1])
-            subprocess.call("cat %s* > %s.zip" % (zip_filepath, zip_filepath), shell=True)
+            subprocess.call(f"cat {zip_filepath}* > {zip_filepath}.zip", shell=True)
             zip_filepath += ".zip"
         extract_path = zip_filepath.strip(".zip")
 
@@ -100,12 +100,12 @@ def download_and_extract(directory, subset, urls):
         with open(zip_filepath, "rb") as f_zip:
             md5 = hashlib.md5(f_zip.read()).hexdigest()
         if md5 != MD5SUM[subset]:
-            raise ValueError("md5sum of %s mismatch" % zip_filepath)
+            raise ValueError(f"md5sum of {zip_filepath} mismatch")
 
         with zipfile.ZipFile(zip_filepath, "r") as zfile:
             zfile.extractall(directory)
             extract_path_ori = os.path.join(directory, zfile.infolist()[0].filename)
-            subprocess.call("mv %s %s" % (extract_path_ori, extract_path), shell=True)
+            subprocess.call(f"mv {extract_path_ori} {extract_path}", shell=True)
     finally:
         # os.remove(zip_filepath)
         pass
@@ -193,7 +193,7 @@ def convert_audio_and_make_label(input_dir, subset, output_dir, output_file):
         writer.writerow(["wav_filename", "wav_length_ms", "speaker_id", "speaker_name"])
         for wav_file in files:
             writer.writerow(wav_file)
-    logger.info(f"Successfully generated csv file {csv_file_path}")
+    logger.info("Successfully generated csv file %s", csv_file_path)
 
 
 def processor(directory, subset, force_process):
diff --git a/TTS/tts/datasets/__init__.py b/TTS/tts/datasets/__init__.py
index a99eb4290d..2f5357c642 100644
--- a/TTS/tts/datasets/__init__.py
+++ b/TTS/tts/datasets/__init__.py
@@ -4,7 +4,6 @@
 from collections import Counter
 from collections.abc import Callable
 from pathlib import Path
-from typing import Dict, List, Tuple, Union
 
 import numpy as np
 
@@ -18,7 +17,7 @@ def split_dataset(items, eval_split_max_size=None, eval_split_size=0.01):
     """Split a dataset into train and eval. Consider speaker distribution in multi-speaker training.
 
     Args:
-        items (List[List]):
+        items (list[list]):
             A list of samples. Each sample is a list of `[audio_path, text, speaker_id]`.
 
         eval_split_max_size (int):
@@ -76,12 +75,12 @@ def load_tts_samples(
     eval_split_max_size=None,
     eval_split_size=0.01,
 ) -> tuple[list[list], list[list]]:
-    """Parse the dataset from the datasets config, load the samples as a List and load the attention alignments if provided.
+    """Parse the dataset from the datasets config, load the samples as a list and load the attention alignments if provided.
     If `formatter` is not None, apply the formatter to the samples else pick the formatter from the available ones based
     on the dataset name.
 
     Args:
-        datasets (List[Dict], Dict): A list of datasets or a single dataset dictionary. If multiple datasets are
+        datasets (list[dict], dict): A list of datasets or a single dataset dictionary. If multiple datasets are
             in the list, they are all merged.
 
         eval_split (bool, optional): If true, create a evaluation split. If an eval split provided explicitly, generate
@@ -100,7 +99,7 @@ def load_tts_samples(
             If > 1, represents the absolute number of evaluation samples. Defaults to 0.01 (1%).
 
     Returns:
-        Tuple[List[List], List[List]: training and evaluation splits of the dataset.
+        tuple[list[list], list[list]: training and evaluation splits of the dataset.
     """
     meta_data_train_all = []
     meta_data_eval_all = [] if eval_split else None
diff --git a/TTS/tts/layers/delightful_tts/conv_layers.py b/TTS/tts/layers/delightful_tts/conv_layers.py
index 588d236852..5cf41d4ff6 100644
--- a/TTS/tts/layers/delightful_tts/conv_layers.py
+++ b/TTS/tts/layers/delightful_tts/conv_layers.py
@@ -50,7 +50,7 @@ def __init__(
         w_init_gain="linear",
         use_weight_norm=False,
     ):
-        super(ConvNorm, self).__init__()  # pylint: disable=super-with-arguments
+        super().__init__()
         if padding is None:
             assert kernel_size % 2 == 1
             padding = int(dilation * (kernel_size - 1) / 2)
@@ -92,7 +92,7 @@ def __init__(
         lstm_type="bilstm",
         use_linear=True,
     ):
-        super(ConvLSTMLinear, self).__init__()  # pylint: disable=super-with-arguments
+        super().__init__()
         self.out_dim = out_dim
         self.lstm_type = lstm_type
         self.use_linear = use_linear
diff --git a/TTS/tts/layers/delightful_tts/networks.py b/TTS/tts/layers/delightful_tts/networks.py
index d0a4adae79..93b65a2a74 100644
--- a/TTS/tts/layers/delightful_tts/networks.py
+++ b/TTS/tts/layers/delightful_tts/networks.py
@@ -51,7 +51,7 @@ def __init__(
         kernel_size=3,
         use_partial_padding=False,  # pylint: disable=unused-argument
     ):
-        super(BottleneckLayer, self).__init__()  # pylint: disable=super-with-arguments
+        super().__init__()
 
         self.reduction_factor = reduction_factor
         reduced_dim = int(in_dim / reduction_factor)
@@ -194,7 +194,7 @@ class STL(nn.Module):
     """
 
     def __init__(self, n_hidden: int, token_num: int):
-        super(STL, self).__init__()  # pylint: disable=super-with-arguments
+        super().__init__()
 
         num_heads = 1
         E = n_hidden
diff --git a/TTS/tts/layers/tortoise/diffusion.py b/TTS/tts/layers/tortoise/diffusion.py
index 2b29091b44..cfb8fa800d 100644
--- a/TTS/tts/layers/tortoise/diffusion.py
+++ b/TTS/tts/layers/tortoise/diffusion.py
@@ -653,7 +653,7 @@ def p_sample_loop_progressive(
         """
         if device is None:
             device = next(model.parameters()).device
-        assert isinstance(shape, (tuple, list))
+        assert isinstance(shape, tuple | list)
         if noise is not None:
             img = noise
         else:
@@ -805,7 +805,7 @@ def ddim_sample_loop_progressive(
         """
         if device is None:
             device = next(model.parameters()).device
-        assert isinstance(shape, (tuple, list))
+        assert isinstance(shape, tuple | list)
         if noise is not None:
             img = noise
         else:
diff --git a/TTS/tts/layers/tortoise/vocoder.py b/TTS/tts/layers/tortoise/vocoder.py
index 6fd784f486..e7497d8190 100644
--- a/TTS/tts/layers/tortoise/vocoder.py
+++ b/TTS/tts/layers/tortoise/vocoder.py
@@ -293,7 +293,7 @@ def __init__(
         hop_length=256,
         n_mel_channels=100,
     ):
-        super(UnivNetGenerator, self).__init__()
+        super().__init__()
         self.mel_channel = n_mel_channels
         self.noise_dim = noise_dim
         self.hop_length = hop_length
@@ -344,7 +344,7 @@ def forward(self, c, z):
         return z
 
     def eval(self, inference=False):
-        super(UnivNetGenerator, self).eval()
+        super().eval()
         # don't remove weight norm while validation in training loop
         if inference:
             self.remove_weight_norm()
diff --git a/TTS/tts/layers/xtts/tokenizer.py b/TTS/tts/layers/xtts/tokenizer.py
index 8af4e78a40..ef4162a1cb 100644
--- a/TTS/tts/layers/xtts/tokenizer.py
+++ b/TTS/tts/layers/xtts/tokenizer.py
@@ -76,7 +76,7 @@ def split_sentence(text, lang, text_split_length=250):
 # List of (regular expression, replacement) pairs for abbreviations:
 _abbreviations = {
     "en": [
-        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        (re.compile(f"\\b{x[0]}\\.", re.IGNORECASE), x[1])
         for x in [
             ("mrs", "misess"),
             ("mr", "mister"),
@@ -99,7 +99,7 @@ def split_sentence(text, lang, text_split_length=250):
         ]
     ],
     "es": [
-        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        (re.compile(f"\\b{x[0]}\\.", re.IGNORECASE), x[1])
         for x in [
             ("sra", "señora"),
             ("sr", "señor"),
@@ -112,7 +112,7 @@ def split_sentence(text, lang, text_split_length=250):
         ]
     ],
     "fr": [
-        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        (re.compile(f"\\b{x[0]}\\.", re.IGNORECASE), x[1])
         for x in [
             ("mme", "madame"),
             ("mr", "monsieur"),
@@ -124,7 +124,7 @@ def split_sentence(text, lang, text_split_length=250):
         ]
     ],
     "de": [
-        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        (re.compile(f"\\b{x[0]}\\.", re.IGNORECASE), x[1])
         for x in [
             ("fr", "frau"),
             ("dr", "doktor"),
@@ -134,7 +134,7 @@ def split_sentence(text, lang, text_split_length=250):
         ]
     ],
     "pt": [
-        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        (re.compile(f"\\b{x[0]}\\.", re.IGNORECASE), x[1])
         for x in [
             ("sra", "senhora"),
             ("sr", "senhor"),
@@ -147,7 +147,7 @@ def split_sentence(text, lang, text_split_length=250):
         ]
     ],
     "it": [
-        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        (re.compile(f"\\b{x[0]}\\.", re.IGNORECASE), x[1])
         for x in [
             # ("sig.ra", "signora"),
             ("sig", "signore"),
@@ -159,7 +159,7 @@ def split_sentence(text, lang, text_split_length=250):
         ]
     ],
     "pl": [
-        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        (re.compile(f"\\b{x[0]}\\.", re.IGNORECASE), x[1])
         for x in [
             ("p", "pani"),
             ("m", "pan"),
@@ -169,19 +169,19 @@ def split_sentence(text, lang, text_split_length=250):
         ]
     ],
     "ar": [
-        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        (re.compile(f"\\b{x[0]}\\.", re.IGNORECASE), x[1])
         for x in [
             # There are not many common abbreviations in Arabic as in English.
         ]
     ],
     "zh": [
-        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        (re.compile(f"\\b{x[0]}\\.", re.IGNORECASE), x[1])
         for x in [
             # Chinese doesn't typically use abbreviations in the same way as Latin-based scripts.
         ]
     ],
     "cs": [
-        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        (re.compile(f"\\b{x[0]}\\.", re.IGNORECASE), x[1])
         for x in [
             ("dr", "doktor"),  # doctor
             ("ing", "inženýr"),  # engineer
@@ -190,7 +190,7 @@ def split_sentence(text, lang, text_split_length=250):
         ]
     ],
     "ru": [
-        (re.compile("\\b%s\\b" % x[0], re.IGNORECASE), x[1])
+        (re.compile(f"\\b{x[0]}\\b", re.IGNORECASE), x[1])
         for x in [
             ("г-жа", "госпожа"),  # Mrs.
             ("г-н", "господин"),  # Mr.
@@ -199,7 +199,7 @@ def split_sentence(text, lang, text_split_length=250):
         ]
     ],
     "nl": [
-        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        (re.compile(f"\\b{x[0]}\\.", re.IGNORECASE), x[1])
         for x in [
             ("dhr", "de heer"),  # Mr.
             ("mevr", "mevrouw"),  # Mrs.
@@ -209,7 +209,7 @@ def split_sentence(text, lang, text_split_length=250):
         ]
     ],
     "tr": [
-        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        (re.compile(f"\\b{x[0]}\\.", re.IGNORECASE), x[1])
         for x in [
             ("b", "bay"),  # Mr.
             ("byk", "büyük"),  # büyük
@@ -218,7 +218,7 @@ def split_sentence(text, lang, text_split_length=250):
         ]
     ],
     "hu": [
-        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        (re.compile(f"\\b{x[0]}\\.", re.IGNORECASE), x[1])
         for x in [
             ("dr", "doktor"),  # doctor
             ("b", "bácsi"),  # Mr.
@@ -227,13 +227,13 @@ def split_sentence(text, lang, text_split_length=250):
         ]
     ],
     "ko": [
-        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        (re.compile(f"\\b{x[0]}\\.", re.IGNORECASE), x[1])
         for x in [
             # Korean doesn't typically use abbreviations in the same way as Latin-based scripts.
         ]
     ],
     "hi": [
-        (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+        (re.compile(f"\\b{x[0]}\\.", re.IGNORECASE), x[1])
         for x in [
             # Hindi doesn't typically use abbreviations in the same way as Latin-based scripts.
         ]
@@ -249,7 +249,7 @@ def expand_abbreviations_multilingual(text, lang="en"):
 
 _symbols_multilingual = {
     "en": [
-        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        (re.compile(rf"{re.escape(x[0])}", re.IGNORECASE), x[1])
         for x in [
             ("&", " and "),
             ("@", " at "),
@@ -261,7 +261,7 @@ def expand_abbreviations_multilingual(text, lang="en"):
         ]
     ],
     "es": [
-        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        (re.compile(rf"{re.escape(x[0])}", re.IGNORECASE), x[1])
         for x in [
             ("&", " y "),
             ("@", " arroba "),
@@ -273,7 +273,7 @@ def expand_abbreviations_multilingual(text, lang="en"):
         ]
     ],
     "fr": [
-        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        (re.compile(rf"{re.escape(x[0])}", re.IGNORECASE), x[1])
         for x in [
             ("&", " et "),
             ("@", " arobase "),
@@ -285,7 +285,7 @@ def expand_abbreviations_multilingual(text, lang="en"):
         ]
     ],
     "de": [
-        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        (re.compile(rf"{re.escape(x[0])}", re.IGNORECASE), x[1])
         for x in [
             ("&", " und "),
             ("@", " at "),
@@ -297,7 +297,7 @@ def expand_abbreviations_multilingual(text, lang="en"):
         ]
     ],
     "pt": [
-        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        (re.compile(rf"{re.escape(x[0])}", re.IGNORECASE), x[1])
         for x in [
             ("&", " e "),
             ("@", " arroba "),
@@ -309,7 +309,7 @@ def expand_abbreviations_multilingual(text, lang="en"):
         ]
     ],
     "it": [
-        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        (re.compile(rf"{re.escape(x[0])}", re.IGNORECASE), x[1])
         for x in [
             ("&", " e "),
             ("@", " chiocciola "),
@@ -321,7 +321,7 @@ def expand_abbreviations_multilingual(text, lang="en"):
         ]
     ],
     "pl": [
-        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        (re.compile(rf"{re.escape(x[0])}", re.IGNORECASE), x[1])
         for x in [
             ("&", " i "),
             ("@", " małpa "),
@@ -334,7 +334,7 @@ def expand_abbreviations_multilingual(text, lang="en"):
     ],
     "ar": [
         # Arabic
-        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        (re.compile(rf"{re.escape(x[0])}", re.IGNORECASE), x[1])
         for x in [
             ("&", " و "),
             ("@", " على "),
@@ -347,7 +347,7 @@ def expand_abbreviations_multilingual(text, lang="en"):
     ],
     "zh": [
         # Chinese
-        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        (re.compile(rf"{re.escape(x[0])}", re.IGNORECASE), x[1])
         for x in [
             ("&", " 和 "),
             ("@", " 在 "),
@@ -360,7 +360,7 @@ def expand_abbreviations_multilingual(text, lang="en"):
     ],
     "cs": [
         # Czech
-        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        (re.compile(rf"{re.escape(x[0])}", re.IGNORECASE), x[1])
         for x in [
             ("&", " a "),
             ("@", " na "),
@@ -373,7 +373,7 @@ def expand_abbreviations_multilingual(text, lang="en"):
     ],
     "ru": [
         # Russian
-        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        (re.compile(rf"{re.escape(x[0])}", re.IGNORECASE), x[1])
         for x in [
             ("&", " и "),
             ("@", " собака "),
@@ -386,7 +386,7 @@ def expand_abbreviations_multilingual(text, lang="en"):
     ],
     "nl": [
         # Dutch
-        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        (re.compile(rf"{re.escape(x[0])}", re.IGNORECASE), x[1])
         for x in [
             ("&", " en "),
             ("@", " bij "),
@@ -398,7 +398,7 @@ def expand_abbreviations_multilingual(text, lang="en"):
         ]
     ],
     "tr": [
-        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        (re.compile(rf"{re.escape(x[0])}", re.IGNORECASE), x[1])
         for x in [
             ("&", " ve "),
             ("@", " at "),
@@ -410,7 +410,7 @@ def expand_abbreviations_multilingual(text, lang="en"):
         ]
     ],
     "hu": [
-        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        (re.compile(rf"{re.escape(x[0])}", re.IGNORECASE), x[1])
         for x in [
             ("&", " és "),
             ("@", " kukac "),
@@ -423,7 +423,7 @@ def expand_abbreviations_multilingual(text, lang="en"):
     ],
     "ko": [
         # Korean
-        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        (re.compile(rf"{re.escape(x[0])}", re.IGNORECASE), x[1])
         for x in [
             ("&", " 그리고 "),
             ("@", " 에 "),
@@ -435,7 +435,7 @@ def expand_abbreviations_multilingual(text, lang="en"):
         ]
     ],
     "hi": [
-        (re.compile(r"%s" % re.escape(x[0]), re.IGNORECASE), x[1])
+        (re.compile(rf"{re.escape(x[0])}", re.IGNORECASE), x[1])
         for x in [
             ("&", " और "),
             ("@", " ऐट दी रेट "),
diff --git a/TTS/tts/layers/xtts/trainer/gpt_trainer.py b/TTS/tts/layers/xtts/trainer/gpt_trainer.py
index 757d1649c2..e00ce2b4de 100644
--- a/TTS/tts/layers/xtts/trainer/gpt_trainer.py
+++ b/TTS/tts/layers/xtts/trainer/gpt_trainer.py
@@ -430,7 +430,7 @@ def get_optimizer(self) -> list:
                     v.is_norm = isinstance(m, norm_modules)
                     v.is_emb = isinstance(m, emb_modules)
 
-                    fpn = "%s.%s" % (mn, k) if mn else k  # full param name
+                    fpn = f"{mn}.{k}" if mn else k  # full param name
                     all_param_names.add(fpn)
                     param_map[fpn] = v
                     if v.is_bias or v.is_norm or v.is_emb:
diff --git a/TTS/tts/layers/xtts/zh_num2words.py b/TTS/tts/layers/xtts/zh_num2words.py
index f0a1db786d..360d9b06c8 100644
--- a/TTS/tts/layers/xtts/zh_num2words.py
+++ b/TTS/tts/layers/xtts/zh_num2words.py
@@ -420,7 +420,7 @@ class ChineseNumberUnit(ChineseChar):
     """
 
     def __init__(self, power, simplified, traditional, big_s, big_t):
-        super(ChineseNumberUnit, self).__init__(simplified, traditional)
+        super().__init__(simplified, traditional)
         self.power = power
         self.big_s = big_s
         self.big_t = big_t
@@ -456,7 +456,7 @@ class ChineseNumberDigit(ChineseChar):
     """
 
     def __init__(self, value, simplified, traditional, big_s, big_t, alt_s=None, alt_t=None):
-        super(ChineseNumberDigit, self).__init__(simplified, traditional)
+        super().__init__(simplified, traditional)
         self.value = value
         self.big_s = big_s
         self.big_t = big_t
@@ -477,7 +477,7 @@ class ChineseMath(ChineseChar):
     """
 
     def __init__(self, simplified, traditional, symbol, expression=None):
-        super(ChineseMath, self).__init__(simplified, traditional)
+        super().__init__(simplified, traditional)
         self.symbol = symbol
         self.expression = expression
         self.big_s = simplified
@@ -507,8 +507,7 @@ def __init__(self, positive, negative, point):
         self.point = point
 
     def __iter__(self):
-        for v in self.__dict__.values():
-            yield v
+        yield from self.__dict__.values()
 
 
 # class OtherSymbol(object):
@@ -702,7 +701,7 @@ def get_value(value_string, use_zeros=True):
             if isinstance(v, CND) and v.value == 2:
                 next_symbol = result_symbols[i + 1] if i < len(result_symbols) - 1 else None
                 previous_symbol = result_symbols[i - 1] if i > 0 else None
-                if isinstance(next_symbol, CNU) and isinstance(previous_symbol, (CNU, type(None))):
+                if isinstance(next_symbol, CNU) and isinstance(previous_symbol, CNU | type(None)):
                     if next_symbol.power != 1 and ((previous_symbol is None) or (previous_symbol.power != 1)):
                         result_symbols[i] = liang
 
diff --git a/TTS/tts/models/__init__.py b/TTS/tts/models/__init__.py
index 0e1587b60b..4746b13ea2 100644
--- a/TTS/tts/models/__init__.py
+++ b/TTS/tts/models/__init__.py
@@ -1,5 +1,4 @@
 import logging
-from typing import Dict, List, Union
 
 from TTS.utils.generic_utils import find_module
 
diff --git a/TTS/tts/utils/text/cmudict.py b/TTS/tts/utils/text/cmudict.py
index 041b42ff31..9c0df06196 100644
--- a/TTS/tts/utils/text/cmudict.py
+++ b/TTS/tts/utils/text/cmudict.py
@@ -119,7 +119,7 @@ def get_arpabet(word, cmudict, punctuation_symbols):
             word = word[:-1]
         arpabet = cmudict.lookup(word)
         if arpabet is not None:
-            return first_symbol + "{%s}" % arpabet[0] + last_symbol
+            return first_symbol + "{%s}" % arpabet[0] + last_symbol  # noqa: UP031
         return first_symbol + word + last_symbol
 
 
diff --git a/TTS/tts/utils/text/english/abbreviations.py b/TTS/tts/utils/text/english/abbreviations.py
index cd93c13c8e..20042b255b 100644
--- a/TTS/tts/utils/text/english/abbreviations.py
+++ b/TTS/tts/utils/text/english/abbreviations.py
@@ -2,7 +2,7 @@
 
 # List of (regular expression, replacement) pairs for abbreviations in english:
 abbreviations_en = [
-    (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+    (re.compile(f"\\b{x[0]}\\.", re.IGNORECASE), x[1])
     for x in [
         ("mrs", "misess"),
         ("mr", "mister"),
diff --git a/TTS/tts/utils/text/french/abbreviations.py b/TTS/tts/utils/text/french/abbreviations.py
index f580dfed7b..e317bbbf3a 100644
--- a/TTS/tts/utils/text/french/abbreviations.py
+++ b/TTS/tts/utils/text/french/abbreviations.py
@@ -2,7 +2,7 @@
 
 # List of (regular expression, replacement) pairs for abbreviations in french:
 abbreviations_fr = [
-    (re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
+    (re.compile(f"\\b{x[0]}\\.", re.IGNORECASE), x[1])
     for x in [
         ("M", "monsieur"),
         ("Mlle", "mademoiselle"),
@@ -38,7 +38,7 @@
         ("boul", "boulevard"),
     ]
 ] + [
-    (re.compile("\\b%s" % x[0]), x[1])
+    (re.compile(f"\\b{x[0]}"), x[1])
     for x in [
         ("Mlle", "mademoiselle"),
         ("Mlles", "mesdemoiselles"),
diff --git a/TTS/tts/utils/text/phonemizers/espeak_wrapper.py b/TTS/tts/utils/text/phonemizers/espeak_wrapper.py
index 5c58afdf08..dbcb8994a7 100644
--- a/TTS/tts/utils/text/phonemizers/espeak_wrapper.py
+++ b/TTS/tts/utils/text/phonemizers/espeak_wrapper.py
@@ -183,7 +183,7 @@ def phonemize_espeak(self, text: str, separator: str = "|", *, tie: bool = False
             else:
                 args.append("--ipa=1")
         if tie:
-            args.append("--tie=%s" % tie)
+            args.append(f"--tie={tie}")
 
         tmp = tempfile.NamedTemporaryFile(mode="w+t", delete=False, encoding="utf8")
         tmp.write(text)
diff --git a/TTS/tts/utils/text/tokenizer.py b/TTS/tts/utils/text/tokenizer.py
index 4d6c9e401e..07a8753884 100644
--- a/TTS/tts/utils/text/tokenizer.py
+++ b/TTS/tts/utils/text/tokenizer.py
@@ -164,7 +164,7 @@ def init_from_config(config: "Coqpit", characters: "BaseCharacters" = None):
         """
         # init cleaners
         text_cleaner = None
-        if isinstance(config.text_cleaner, (str, list)):
+        if isinstance(config.text_cleaner, str | list):
             text_cleaner = getattr(cleaners, config.text_cleaner)
 
         # init characters
diff --git a/TTS/utils/radam.py b/TTS/utils/radam.py
index b5306d6ab3..b893d115c9 100644
--- a/TTS/utils/radam.py
+++ b/TTS/utils/radam.py
@@ -18,7 +18,7 @@ def __init__(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-8, weight_decay=0
             raise ValueError(f"Invalid beta parameter at index 1: {betas[1]}")
 
         self.degenerated_to_sgd = degenerated_to_sgd
-        if isinstance(params, (list, tuple)) and len(params) > 0 and isinstance(params[0], dict):
+        if isinstance(params, list | tuple) and len(params) > 0 and isinstance(params[0], dict):
             for param in params:
                 if "betas" in param and (param["betas"][0] != betas[0] or param["betas"][1] != betas[1]):
                     param["buffer"] = [[None, None, None] for _ in range(10)]
diff --git a/TTS/vc/layers/freevc/modules.py b/TTS/vc/layers/freevc/modules.py
index c34f22d701..92df39b5e0 100644
--- a/TTS/vc/layers/freevc/modules.py
+++ b/TTS/vc/layers/freevc/modules.py
@@ -48,7 +48,7 @@ def forward(self, x, x_mask):
 
 class WN(torch.nn.Module):
     def __init__(self, hidden_channels, kernel_size, dilation_rate, n_layers, gin_channels=0, p_dropout=0):
-        super(WN, self).__init__()
+        super().__init__()
         assert kernel_size % 2 == 1
         self.hidden_channels = hidden_channels
         self.kernel_size = (kernel_size,)
@@ -122,7 +122,7 @@ def remove_weight_norm(self):
 
 class ResBlock1(torch.nn.Module):
     def __init__(self, channels, kernel_size=3, dilation=(1, 3, 5)):
-        super(ResBlock1, self).__init__()
+        super().__init__()
         self.convs1 = nn.ModuleList(
             [
                 weight_norm(
@@ -198,7 +198,7 @@ def remove_weight_norm(self):
 
 class ResBlock2(torch.nn.Module):
     def __init__(self, channels, kernel_size=3, dilation=(1, 3)):
-        super(ResBlock2, self).__init__()
+        super().__init__()
         self.convs = nn.ModuleList(
             [
                 weight_norm(
diff --git a/TTS/vc/layers/freevc/wavlm/modules.py b/TTS/vc/layers/freevc/wavlm/modules.py
index 06348e4bb4..cddacd69ab 100644
--- a/TTS/vc/layers/freevc/wavlm/modules.py
+++ b/TTS/vc/layers/freevc/wavlm/modules.py
@@ -88,7 +88,7 @@ class Swish(nn.Module):
 
     def __init__(self):
         """Construct an MultiHeadedAttention object."""
-        super(Swish, self).__init__()
+        super().__init__()
         self.act = torch.nn.Sigmoid()
 
     def forward(self, x):
@@ -97,7 +97,7 @@ def forward(self, x):
 
 class GLU_Linear(nn.Module):
     def __init__(self, input_dim, output_dim, glu_type="sigmoid", bias_in_glu=True):
-        super(GLU_Linear, self).__init__()
+        super().__init__()
 
         self.glu_type = glu_type
         self.output_dim = output_dim
@@ -218,7 +218,7 @@ def quant_noise(module, p, block_size):
         return module
 
     # supported modules
-    assert isinstance(module, (nn.Linear, nn.Embedding, nn.Conv2d))
+    assert isinstance(module, nn.Linear | nn.Embedding | nn.Conv2d)
 
     # test whether module.weight has the right sizes wrt block_size
     is_conv = module.weight.ndim == 4
diff --git a/TTS/vc/models/__init__.py b/TTS/vc/models/__init__.py
index 138178089f..4a1833bf50 100644
--- a/TTS/vc/models/__init__.py
+++ b/TTS/vc/models/__init__.py
@@ -1,7 +1,6 @@
 import importlib
 import logging
 import re
-from typing import Dict, List, Union
 
 logger = logging.getLogger(__name__)
 
diff --git a/TTS/vc/models/freevc.py b/TTS/vc/models/freevc.py
index 69fa40bdbc..7f132772ea 100644
--- a/TTS/vc/models/freevc.py
+++ b/TTS/vc/models/freevc.py
@@ -101,7 +101,7 @@ def __init__(
         upsample_kernel_sizes,
         gin_channels=0,
     ):
-        super(Generator, self).__init__()
+        super().__init__()
         self.num_kernels = len(resblock_kernel_sizes)
         self.num_upsamples = len(upsample_rates)
         self.conv_pre = Conv1d(initial_channel, upsample_initial_channel, 7, 1, padding=3)
@@ -164,7 +164,7 @@ def remove_weight_norm(self):
 
 class MultiPeriodDiscriminator(torch.nn.Module):
     def __init__(self, use_spectral_norm=False):
-        super(MultiPeriodDiscriminator, self).__init__()
+        super().__init__()
         periods = [2, 3, 5, 7, 11]
 
         discs = [DiscriminatorS(use_spectral_norm=use_spectral_norm)]
@@ -189,7 +189,7 @@ def forward(self, y, y_hat):
 
 class SpeakerEncoder(torch.nn.Module):
     def __init__(self, mel_n_channels=80, model_num_layers=3, model_hidden_size=256, model_embedding_size=256):
-        super(SpeakerEncoder, self).__init__()
+        super().__init__()
         self.lstm = nn.LSTM(mel_n_channels, model_hidden_size, model_num_layers, batch_first=True)
         self.linear = nn.Linear(model_hidden_size, model_embedding_size)
         self.relu = nn.ReLU()
diff --git a/TTS/vocoder/datasets/__init__.py b/TTS/vocoder/datasets/__init__.py
index d935209348..cef6a50b05 100644
--- a/TTS/vocoder/datasets/__init__.py
+++ b/TTS/vocoder/datasets/__init__.py
@@ -1,5 +1,3 @@
-from typing import List
-
 from coqpit import Coqpit
 from torch.utils.data import Dataset
 
diff --git a/TTS/vocoder/datasets/gan_dataset.py b/TTS/vocoder/datasets/gan_dataset.py
index 0806c0d496..c0882c701f 100644
--- a/TTS/vocoder/datasets/gan_dataset.py
+++ b/TTS/vocoder/datasets/gan_dataset.py
@@ -32,7 +32,7 @@ def __init__(
         super().__init__()
         self.ap = ap
         self.item_list = items
-        self.compute_feat = not isinstance(items[0], (tuple, list))
+        self.compute_feat = not isinstance(items[0], tuple | list)
         self.seq_len = seq_len
         self.hop_len = hop_len
         self.pad_short = pad_short
diff --git a/TTS/vocoder/datasets/wavernn_dataset.py b/TTS/vocoder/datasets/wavernn_dataset.py
index 4c4f5c48df..ffb71177c5 100644
--- a/TTS/vocoder/datasets/wavernn_dataset.py
+++ b/TTS/vocoder/datasets/wavernn_dataset.py
@@ -18,7 +18,7 @@ class WaveRNNDataset(Dataset):
     def __init__(self, ap, items, seq_len, hop_len, pad, mode, mulaw, is_training=True, return_segments=True):
         super().__init__()
         self.ap = ap
-        self.compute_feat = not isinstance(items[0], (tuple, list))
+        self.compute_feat = not isinstance(items[0], tuple | list)
         self.item_list = items
         self.seq_len = seq_len
         self.hop_len = hop_len
diff --git a/TTS/vocoder/layers/wavegrad.py b/TTS/vocoder/layers/wavegrad.py
index 9f1512c6d4..187e7062e2 100644
--- a/TTS/vocoder/layers/wavegrad.py
+++ b/TTS/vocoder/layers/wavegrad.py
@@ -74,7 +74,7 @@ def shif_and_scale(x, scale, shift):
 class UBlock(nn.Module):
     def __init__(self, input_size, hidden_size, factor, dilation):
         super().__init__()
-        assert isinstance(dilation, (list, tuple))
+        assert isinstance(dilation, list | tuple)
         assert len(dilation) == 4
 
         self.factor = factor
diff --git a/TTS/vocoder/models/parallel_wavegan_discriminator.py b/TTS/vocoder/models/parallel_wavegan_discriminator.py
index 211d45d91c..02ad60e0ff 100644
--- a/TTS/vocoder/models/parallel_wavegan_discriminator.py
+++ b/TTS/vocoder/models/parallel_wavegan_discriminator.py
@@ -71,7 +71,7 @@ def forward(self, x):
 
     def apply_weight_norm(self):
         def _apply_weight_norm(m):
-            if isinstance(m, (torch.nn.Conv1d, torch.nn.Conv2d)):
+            if isinstance(m, torch.nn.Conv1d | torch.nn.Conv2d):
                 torch.nn.utils.parametrizations.weight_norm(m)
 
         self.apply(_apply_weight_norm)
@@ -174,7 +174,7 @@ def forward(self, x):
 
     def apply_weight_norm(self):
         def _apply_weight_norm(m):
-            if isinstance(m, (torch.nn.Conv1d, torch.nn.Conv2d)):
+            if isinstance(m, torch.nn.Conv1d | torch.nn.Conv2d):
                 torch.nn.utils.parametrizations.weight_norm(m)
 
         self.apply(_apply_weight_norm)
diff --git a/TTS/vocoder/models/parallel_wavegan_generator.py b/TTS/vocoder/models/parallel_wavegan_generator.py
index 0659a00cc1..f4ef3a0734 100644
--- a/TTS/vocoder/models/parallel_wavegan_generator.py
+++ b/TTS/vocoder/models/parallel_wavegan_generator.py
@@ -145,7 +145,7 @@ def _remove_weight_norm(m):
 
     def apply_weight_norm(self):
         def _apply_weight_norm(m):
-            if isinstance(m, (torch.nn.Conv1d, torch.nn.Conv2d)):
+            if isinstance(m, torch.nn.Conv1d | torch.nn.Conv2d):
                 torch.nn.utils.parametrizations.weight_norm(m)
                 logger.info("Weight norm is applied to %s", m)
 
diff --git a/TTS/vocoder/models/univnet_generator.py b/TTS/vocoder/models/univnet_generator.py
index 82909231ee..d991941441 100644
--- a/TTS/vocoder/models/univnet_generator.py
+++ b/TTS/vocoder/models/univnet_generator.py
@@ -127,7 +127,7 @@ def apply_weight_norm(self):
         """Apply weight normalization module from all of the layers."""
 
         def _apply_weight_norm(m):
-            if isinstance(m, (torch.nn.Conv1d, torch.nn.Conv2d)):
+            if isinstance(m, torch.nn.Conv1d | torch.nn.Conv2d):
                 torch.nn.utils.parametrizations.weight_norm(m)
                 logger.info("Weight norm is applied to %s", m)
 
diff --git a/tests/tts_tests/test_tacotron_model.py b/tests/tts_tests/test_tacotron_model.py
index e71baa574d..3976b9ae8d 100644
--- a/tests/tts_tests/test_tacotron_model.py
+++ b/tests/tts_tests/test_tacotron_model.py
@@ -51,7 +51,7 @@ def test_train_step():
         criterion_st = nn.BCEWithLogitsLoss().to(device)
         model = Tacotron(config).to(device)  # FIXME: missing num_speakers parameter to Tacotron ctor
         model.train()
-        print(" > Num parameters for Tacotron model:%s" % (count_parameters(model)))
+        print(f" > Num parameters for Tacotron model:{count_parameters(model)}")
         model_ref = copy.deepcopy(model)
         count = 0
         for param, param_ref in zip(model.parameters(), model_ref.parameters()):
@@ -105,7 +105,7 @@ def test_train_step():
         config.d_vector_dim = 55
         model = Tacotron(config).to(device)  # FIXME: missing num_speakers parameter to Tacotron ctor
         model.train()
-        print(" > Num parameters for Tacotron model:%s" % (count_parameters(model)))
+        print(f" > Num parameters for Tacotron model:{count_parameters(model)}")
         model_ref = copy.deepcopy(model)
         count = 0
         for param, param_ref in zip(model.parameters(), model_ref.parameters()):
@@ -165,7 +165,7 @@ def test_train_step():
         model = Tacotron(config).to(device)  # FIXME: missing num_speakers parameter to Tacotron ctor
         model.train()
         # print(model)
-        print(" > Num parameters for Tacotron GST model:%s" % (count_parameters(model)))
+        print(f" > Num parameters for Tacotron GST model:{count_parameters(model)}")
         model_ref = copy.deepcopy(model)
         count = 0
         for param, param_ref in zip(model.parameters(), model_ref.parameters()):
@@ -217,7 +217,7 @@ def test_train_step():
         model = Tacotron(config).to(device)  # FIXME: missing num_speakers parameter to Tacotron ctor
         model.train()
         # print(model)
-        print(" > Num parameters for Tacotron GST model:%s" % (count_parameters(model)))
+        print(f" > Num parameters for Tacotron GST model:{count_parameters(model)}")
         model_ref = copy.deepcopy(model)
         count = 0
         for param, param_ref in zip(model.parameters(), model_ref.parameters()):
@@ -288,7 +288,7 @@ def test_train_step():
         criterion = model.get_criterion()
         optimizer = model.get_optimizer()
         model.train()
-        print(" > Num parameters for Tacotron with Capacitron VAE model:%s" % (count_parameters(model)))
+        print(f" > Num parameters for Tacotron with Capacitron VAE model:{count_parameters(model)}")
         model_ref = copy.deepcopy(model)
         count = 0
         for param, param_ref in zip(model.parameters(), model_ref.parameters()):
@@ -341,7 +341,7 @@ def test_train_step():
         config.d_vector_dim = 55
         model = Tacotron(config).to(device)  # FIXME: missing num_speakers parameter to Tacotron ctor
         model.train()
-        print(" > Num parameters for Tacotron model:%s" % (count_parameters(model)))
+        print(f" > Num parameters for Tacotron model:{count_parameters(model)}")
         model_ref = copy.deepcopy(model)
         count = 0
         for param, param_ref in zip(model.parameters(), model_ref.parameters()):
diff --git a/tests/tts_tests2/test_glow_tts.py b/tests/tts_tests2/test_glow_tts.py
index d0557b6311..967e9ecb9e 100644
--- a/tests/tts_tests2/test_glow_tts.py
+++ b/tests/tts_tests2/test_glow_tts.py
@@ -107,7 +107,7 @@ def _test_forward(self, batch_size):
         config = GlowTTSConfig(num_chars=32)
         model = GlowTTS(config).to(device)
         model.train()
-        print(" > Num parameters for GlowTTS model:%s" % (count_parameters(model)))
+        print(f" > Num parameters for GlowTTS model:{count_parameters(model)}")
         # inference encoder and decoder with MAS
         y = model.forward(input_dummy, input_lengths, mel_spec, mel_lengths)
         self.assertEqual(y["z"].shape, mel_spec.shape)
@@ -134,7 +134,7 @@ def _test_forward_with_d_vector(self, batch_size):
         )
         model = GlowTTS.init_from_config(config).to(device)
         model.train()
-        print(" > Num parameters for GlowTTS model:%s" % (count_parameters(model)))
+        print(f" > Num parameters for GlowTTS model:{count_parameters(model)}")
         # inference encoder and decoder with MAS
         y = model.forward(input_dummy, input_lengths, mel_spec, mel_lengths, {"d_vectors": d_vector})
         self.assertEqual(y["z"].shape, mel_spec.shape)
@@ -160,7 +160,7 @@ def _test_forward_with_speaker_id(self, batch_size):
         )
         model = GlowTTS.init_from_config(config).to(device)
         model.train()
-        print(" > Num parameters for GlowTTS model:%s" % (count_parameters(model)))
+        print(f" > Num parameters for GlowTTS model:{count_parameters(model)}")
         # inference encoder and decoder with MAS
         y = model.forward(input_dummy, input_lengths, mel_spec, mel_lengths, {"speaker_ids": speaker_ids})
         self.assertEqual(y["z"].shape, mel_spec.shape)
@@ -261,7 +261,7 @@ def test_train_step(self):
         # reference model to compare model weights
         model_ref = GlowTTS(config).to(device)
         model.train()
-        print(" > Num parameters for GlowTTS model:%s" % (count_parameters(model)))
+        print(f" > Num parameters for GlowTTS model:{count_parameters(model)}")
         # pass the state to ref model
         model_ref.load_state_dict(copy.deepcopy(model.state_dict()))
         count = 0
diff --git a/tests/vc_tests/test_freevc.py b/tests/vc_tests/test_freevc.py
index fe07b2723c..dd45d6941f 100644
--- a/tests/vc_tests/test_freevc.py
+++ b/tests/vc_tests/test_freevc.py
@@ -55,7 +55,7 @@ def _test_forward(self, batch_size):
         config = FreeVCConfig()
         model = FreeVC(config).to(device)
         model.train()
-        print(" > Num parameters for FreeVC model:%s" % (count_parameters(model)))
+        print(f" > Num parameters for FreeVC model:{count_parameters(model)}")
 
         mel, spec, spec_lengths, waveform = self._create_inputs(config, batch_size)