more code refactor

GPT_SoVITS/text/japanese.py

@@ -8,57 +8,63 @@ from text import symbols
 
 # Regular expression matching Japanese without punctuation marks:
 _japanese_characters = re.compile(
-    r'[A-Za-z\d\u3005\u3040-\u30ff\u4e00-\u9fff\uff11-\uff19\uff21-\uff3a\uff41-\uff5a\uff66-\uff9d]')
+    r"[A-Za-z\d\u3005\u3040-\u30ff\u4e00-\u9fff\uff11-\uff19\uff21-\uff3a\uff41-\uff5a\uff66-\uff9d]"
+)
 
 # Regular expression matching non-Japanese characters or punctuation marks:
 _japanese_marks = re.compile(
-    r'[^A-Za-z\d\u3005\u3040-\u30ff\u4e00-\u9fff\uff11-\uff19\uff21-\uff3a\uff41-\uff5a\uff66-\uff9d]')
+    r"[^A-Za-z\d\u3005\u3040-\u30ff\u4e00-\u9fff\uff11-\uff19\uff21-\uff3a\uff41-\uff5a\uff66-\uff9d]"
+)
 
 # List of (symbol, Japanese) pairs for marks:
-_symbols_to_japanese = [(re.compile('%s' % x[0]), x[1]) for x in [
-    ('％', 'パーセント')
-]]
+_symbols_to_japanese = [(re.compile("%s" % x[0]), x[1]) for x in [("％", "パーセント")]]
 
 
 # List of (consonant, sokuon) pairs:
-_real_sokuon = [(re.compile('%s' % x[0]), x[1]) for x in [
-    (r'Q([↑↓]*[kg])', r'k#\1'),
-    (r'Q([↑↓]*[tdjʧ])', r't#\1'),
-    (r'Q([↑↓]*[sʃ])', r's\1'),
-    (r'Q([↑↓]*[pb])', r'p#\1')
-]]
+_real_sokuon = [
+    (re.compile("%s" % x[0]), x[1])
+    for x in [
+        (r"Q([↑↓]*[kg])", r"k#\1"),
+        (r"Q([↑↓]*[tdjʧ])", r"t#\1"),
+        (r"Q([↑↓]*[sʃ])", r"s\1"),
+        (r"Q([↑↓]*[pb])", r"p#\1"),
+    ]
+]
 
 # List of (consonant, hatsuon) pairs:
-_real_hatsuon = [(re.compile('%s' % x[0]), x[1]) for x in [
-    (r'N([↑↓]*[pbm])', r'm\1'),
-    (r'N([↑↓]*[ʧʥj])', r'n^\1'),
-    (r'N([↑↓]*[tdn])', r'n\1'),
-    (r'N([↑↓]*[kg])', r'ŋ\1')
-]]
-
+_real_hatsuon = [
+    (re.compile("%s" % x[0]), x[1])
+    for x in [
+        (r"N([↑↓]*[pbm])", r"m\1"),
+        (r"N([↑↓]*[ʧʥj])", r"n^\1"),
+        (r"N([↑↓]*[tdn])", r"n\1"),
+        (r"N([↑↓]*[kg])", r"ŋ\1"),
+    ]
+]
 
 
 def post_replace_ph(ph):
     rep_map = {
-        '：': ',',
-        '；': ',',
-        '，': ',',
-        '。': '.',
-        '！': '!',
-        '？': '?',
-        '\n': '.',
+        "：": ",",
+        "；": ",",
+        "，": ",",
+        "。": ".",
+        "！": "!",
+        "？": "?",
+        "\n": ".",
         "·": ",",
-        '、': ",",
-        '...': '…'
+        "、": ",",
+        "...": "…",
     }
     if ph in rep_map.keys():
         ph = rep_map[ph]
     if ph in symbols:
         return ph
     if ph not in symbols:
-        ph = 'UNK'
+        ph = "UNK"
     return ph
 
+
 def symbols_to_japanese(text):
     for regex, replacement in _symbols_to_japanese:
         text = re.sub(regex, replacement, text)
@@ -66,7 +72,7 @@ def symbols_to_japanese(text):
 
 
 def preprocess_jap(text):
-    '''Reference https://r9y9.github.io/ttslearn/latest/notebooks/ch10_Recipe-Tacotron.html'''
+    """Reference https://r9y9.github.io/ttslearn/latest/notebooks/ch10_Recipe-Tacotron.html"""
     text = symbols_to_japanese(text)
     sentences = re.split(_japanese_marks, text)
     marks = re.findall(_japanese_marks, text)
@@ -77,13 +83,15 @@ def preprocess_jap(text):
             text += p.split(" ")
 
         if i < len(marks):
-            text += [marks[i].replace(' ', '')]
+            text += [marks[i].replace(" ", "")]
     return text
 
+
 def text_normalize(text):
     # todo: jap text normalize
     return text
 
+
 def g2p(norm_text):
     phones = preprocess_jap(norm_text)
     phones = [post_replace_ph(i) for i in phones]
@@ -91,7 +99,7 @@ def g2p(norm_text):
     return phones
 
 
-if __name__ == '__main__':
+if __name__ == "__main__":
     for line in open("../../../Downloads/transcript_utf8.txt").readlines():
         text = line.split(":")[1]
         phones = g2p(text)