fix Lint checks

tests/test_speaker_encoder.py

@@ -6,6 +6,7 @@ from tests import get_tests_input_path
 from TTS.speaker_encoder.losses import AngleProtoLoss, GE2ELoss, SoftmaxAngleProtoLoss
 from TTS.speaker_encoder.models.lstm import LSTMSpeakerEncoder
 from TTS.speaker_encoder.models.resnet import ResNetSpeakerEncoder
+
 file_path = get_tests_input_path()
 
 
@@ -39,6 +40,7 @@ class LSTMSpeakerEncoderTests(unittest.TestCase):
         assert output.shape[1] == 256
         assert len(output.shape) == 2
 
+
 class ResNetSpeakerEncoderTests(unittest.TestCase):
     # pylint: disable=R0201
     def test_in_out(self):
@@ -65,6 +67,7 @@ class ResNetSpeakerEncoderTests(unittest.TestCase):
         assert output.shape[1] == 256
         assert len(output.shape) == 2
 
+
 class GE2ELossTests(unittest.TestCase):
     # pylint: disable=R0201
     def test_in_out(self):
@@ -92,6 +95,7 @@ class GE2ELossTests(unittest.TestCase):
         output = loss.forward(dummy_input)
         assert output.item() < 0.005
 
+
 class AngleProtoLossTests(unittest.TestCase):
     # pylint: disable=R0201
     def test_in_out(self):
@@ -121,6 +125,7 @@ class AngleProtoLossTests(unittest.TestCase):
         output = loss.forward(dummy_input)
         assert output.item() < 0.005
 
+
 class SoftmaxAngleProtoLossTests(unittest.TestCase):
     # pylint: disable=R0201
     def test_in_out(self):

tests/test_speaker_encoder_train.py

@@ -46,7 +46,7 @@ run_cli(command_train)
 shutil.rmtree(continue_path)
 
 # test resnet speaker encoder
-config.model_params['model_name'] = "resnet"
+config.model_params["model_name"] = "resnet"
 config.save_json(config_path)
 
 # train the model for one epoch

tests/text_tests/test_japanese_phonemizer.py

@@ -0,0 +1,24 @@
+import unittest
+
+from TTS.tts.utils.text.japanese.phonemizer import japanese_text_to_phonemes
+
+_TEST_CASES = """
+どちらに行きますか？/dochiraniikimasuka?
+今日は温泉に、行きます。/kyo:waoNseNni,ikimasu.
+「A」から「Z」までです。/AkaraZmadedesu.
+そうですね！/so:desune!
+クジラは哺乳類です。/kujirawahonyu:ruidesu.
+ヴィディオを見ます。/bidioomimasu.
+ky o: w a o N s e N n i , i k i m a s u ./kyo:waoNseNni,ikimasu.
+"""
+
+
+class TestText(unittest.TestCase):
+    def test_japanese_text_to_phonemes(self):
+        for line in _TEST_CASES.strip().split("\n"):
+            text, phone = line.split("/")
+            self.assertEqual(japanese_text_to_phonemes(text), phone)
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/tts_tests/test_glow_tts_train.py

@@ -17,7 +17,7 @@ config = GlowTTSConfig(
     text_cleaner="english_cleaners",
     use_phonemes=True,
     phoneme_language="zh-CN",
-    phoneme_cache_path='tests/data/ljspeech/phoneme_cache/',
+    phoneme_cache_path="tests/data/ljspeech/phoneme_cache/",
     run_eval=True,
     test_delay_epochs=-1,
     epochs=1,

tests/tts_tests/test_speedy_speech_train.py

@@ -17,7 +17,7 @@ config = SpeedySpeechConfig(
     text_cleaner="english_cleaners",
     use_phonemes=True,
     phoneme_language="zh-CN",
-    phoneme_cache_path='tests/data/ljspeech/phoneme_cache/',
+    phoneme_cache_path="tests/data/ljspeech/phoneme_cache/",
     run_eval=True,
     test_delay_epochs=-1,
     epochs=1,

tests/vocoder_tests/test_fullband_melgan_train.py

@@ -20,6 +20,7 @@ config = FullbandMelganConfig(
     eval_split_size=1,
     print_step=1,
     print_eval=True,
+    discriminator_model_params={"base_channels": 16, "max_channels": 256, "downsample_factors": [4, 4, 4]},
     data_path="tests/data/ljspeech",
     output_path=output_path,
 )

tests/vocoder_tests/test_melgan_train.py

@@ -19,6 +19,7 @@ config = MelganConfig(
     seq_len=2048,
     eval_split_size=1,
     print_step=1,
+    discriminator_model_params={"base_channels": 16, "max_channels": 256, "downsample_factors": [4, 4, 4]},
     print_eval=True,
     data_path="tests/data/ljspeech",
     output_path=output_path,