support gpt-sovits v2

support gpt-sovits v2

GPT_SoVITS/text/__init__.py

@@ -1,5 +1,8 @@
-from text.symbols import *
+import os
-
+if os.environ.get("version","v1")=="v1":
+  from text.symbols import symbols
+else:
+  from text.symbols2 import symbols
 
 _symbol_to_id = {s: i for i, s in enumerate(symbols)}
 

GPT_SoVITS/text/cantonese.py

@@ -0,0 +1,204 @@
+import sys
+# sys.path.append("/data/docker/liujing04/gpt-vits/mq-vits-s1bert_no_bert-kv_cache-new_text-emo")
+sys.path.append("/data/docker/liujing04/gpt-vits/mq-vits-s1bert_no_bert-kv_cache-new_text-emo2v2")
+import re
+import cn2an
+
+from pyjyutping import jyutping
+from text.symbols import punctuation
+
+normalizer = lambda x: cn2an.transform(x, "an2cn")
+
+INITIALS = [
+    "aa",
+    "aai",
+    "aak",
+    "aap",
+    "aat",
+    "aau",
+    "ai",
+    "au",
+    "ap",
+    "at",
+    "ak",
+    "a",
+    "p",
+    "b",
+    "e",
+    "ts",
+    "t",
+    "dz",
+    "d",
+    "kw",
+    "k",
+    "gw",
+    "g",
+    "f",
+    "h",
+    "l",
+    "m",
+    "ng",
+    "n",
+    "s",
+    "y",
+    "w",
+    "c",
+    "z",
+    "j",
+    "ong",
+    "on",
+    "ou",
+    "oi",
+    "ok",
+    "o",
+    "uk",
+    "ung",
+]
+INITIALS += ["sp", "spl", "spn", "sil"]
+
+
+rep_map = {
+    "：": ",",
+    "；": ",",
+    "，": ",",
+    "。": ".",
+    "！": "!",
+    "？": "?",
+    "\n": ".",
+    "·": ",",
+    "、": ",",
+    "...": "…",
+    "$": ".",
+    "“": "'",
+    "”": "'",
+    '"': "'",
+    "‘": "'",
+    "’": "'",
+    "（": "'",
+    "）": "'",
+    "(": "'",
+    ")": "'",
+    "《": "'",
+    "》": "'",
+    "【": "'",
+    "】": "'",
+    "[": "'",
+    "]": "'",
+    "—": "-",
+    "～": "-",
+    "~": "-",
+    "「": "'",
+    "」": "'",
+}
+
+
+def replace_punctuation(text):
+    # text = text.replace("嗯", "恩").replace("呣", "母")
+    pattern = re.compile("|".join(re.escape(p) for p in rep_map.keys()))
+
+    replaced_text = pattern.sub(lambda x: rep_map[x.group()], text)
+
+    replaced_text = re.sub(
+        r"[^\u4e00-\u9fa5" + "".join(punctuation) + r"]+", "", replaced_text
+    )
+
+    return replaced_text
+
+
+def text_normalize(text):
+    text = normalizer(text)
+    text = replace_punctuation(text)
+    return text
+
+punctuation_set=set(punctuation)
+def jyuping_to_initials_finals_tones(jyuping_syllables):
+    initials_finals = []
+    tones = []
+    word2ph = []
+
+    for syllable in jyuping_syllables:
+        if syllable in punctuation:
+            initials_finals.append(syllable)
+            tones.append(0)
+            word2ph.append(1)  # Add 1 for punctuation
+        elif syllable == "_":
+            initials_finals.append(syllable)
+            tones.append(0)
+            word2ph.append(1)  # Add 1 for underscore
+        else:
+            try:
+                tone = int(syllable[-1])
+                syllable_without_tone = syllable[:-1]
+            except ValueError:
+                tone = 0
+                syllable_without_tone = syllable
+
+            for initial in INITIALS:
+                if syllable_without_tone.startswith(initial):
+                    if syllable_without_tone.startswith("nga"):
+                        initials_finals.extend(
+                            [
+                                syllable_without_tone[:2],
+                                syllable_without_tone[2:] or syllable_without_tone[-1],
+                            ]
+                        )
+                        # tones.extend([tone, tone])
+                        tones.extend([-1, tone])
+                        word2ph.append(2)
+                    else:
+                        final = syllable_without_tone[len(initial) :] or initial[-1]
+                        initials_finals.extend([initial, final])
+                        # tones.extend([tone, tone])
+                        tones.extend([-1, tone])
+                        word2ph.append(2)
+                    break
+    assert len(initials_finals) == len(tones)
+
+    ###魔改为辅音+带音调的元音
+    phones=[]
+    for a,b in zip(initials_finals,tones):
+        if(b not in [-1,0]):###防止粤语和普通话重合开头加Y，如果是标点，不加。
+            todo="%s%s"%(a,b)
+        else:todo=a
+        if(todo not in punctuation_set):todo="Y%s"%todo
+        phones.append(todo)
+
+    # return initials_finals, tones, word2ph
+    return phones, word2ph
+
+
+def get_jyutping(text):
+    jp = jyutping.convert(text)
+    # print(1111111,jp)
+    for symbol in punctuation:
+        jp = jp.replace(symbol, " " + symbol + " ")
+    jp_array = jp.split()
+    return jp_array
+
+
+def get_bert_feature(text, word2ph):
+    from text import chinese_bert
+
+    return chinese_bert.get_bert_feature(text, word2ph)
+
+
+def g2p(text):
+    # word2ph = []
+    jyuping = get_jyutping(text)
+    # print(jyuping)
+    # phones, tones, word2ph = jyuping_to_initials_finals_tones(jyuping)
+    phones, word2ph = jyuping_to_initials_finals_tones(jyuping)
+    # phones = ["_"] + phones + ["_"]
+    # tones = [0] + tones + [0]
+    # word2ph = [1] + word2ph + [1]
+    return phones, word2ph
+
+
+if __name__ == "__main__":
+    # text = "啊！但是《原神》是由,米哈\游自主，  [研发]的一款全.新开放世界.冒险游戏"
+    text = "佢個鋤頭太短啦。"
+    text = text_normalize(text)
+    # phones, tones, word2ph = g2p(text)
+    phones, word2ph = g2p(text)
+    # print(phones, tones, word2ph)
+    print(phones, word2ph)

GPT_SoVITS/text/chinese2.py

@@ -0,0 +1,187 @@
+import os
+import pdb
+import re
+
+import cn2an
+from pypinyin import lazy_pinyin, Style
+
+from text.symbols import punctuation
+from text.tone_sandhi import ToneSandhi
+from text.zh_normalization.text_normlization import TextNormalizer
+
+normalizer = lambda x: cn2an.transform(x, "an2cn")
+
+current_file_path = os.path.dirname(__file__)
+pinyin_to_symbol_map = {
+    line.split("\t")[0]: line.strip().split("\t")[1]
+    for line in open(os.path.join(current_file_path, "opencpop-strict.txt")).readlines()
+}
+
+import jieba_fast.posseg as psg
+
+# is_g2pw_str = os.environ.get("is_g2pw", "True")##默认开启
+# is_g2pw = False#True if is_g2pw_str.lower() == 'true' else False
+is_g2pw = True#True if is_g2pw_str.lower() == 'true' else False
+if is_g2pw:
+    print("当前使用g2pw进行拼音推理")
+    from text.g2pw import G2PWPinyin
+    parent_directory = os.path.dirname(current_file_path)
+    g2pw = G2PWPinyin(model_dir="GPT_SoVITS/text/G2PWModel",model_source="GPT_SoVITS/pretrained_models/chinese-roberta-wwm-ext-large",v_to_u=False, neutral_tone_with_five=True)
+
+rep_map = {
+    "：": ",",
+    "；": ",",
+    "，": ",",
+    "。": ".",
+    "！": "!",
+    "？": "?",
+    "\n": ".",
+    "·": ",",
+    "、": ",",
+    "...": "…",
+    "$": ".",
+    "/": ",",
+    "—": "-",
+}
+
+tone_modifier = ToneSandhi()
+
+
+def replace_punctuation(text):
+    text = text.replace("嗯", "恩").replace("呣", "母")
+    pattern = re.compile("|".join(re.escape(p) for p in rep_map.keys()))
+
+    replaced_text = pattern.sub(lambda x: rep_map[x.group()], text)
+
+    replaced_text = re.sub(
+        r"[^\u4e00-\u9fa5" + "".join(punctuation) + r"]+", "", replaced_text
+    )
+
+    return replaced_text
+
+
+def g2p(text):
+    pattern = r"(?<=[{0}])\s*".format("".join(punctuation))
+    sentences = [i for i in re.split(pattern, text) if i.strip() != ""]
+    phones, word2ph = _g2p(sentences)
+    return phones, word2ph
+
+
+def _get_initials_finals(word):
+    initials = []
+    finals = []
+
+    if not is_g2pw:
+        orig_initials = lazy_pinyin(word, neutral_tone_with_five=True, style=Style.INITIALS)
+        orig_finals = lazy_pinyin(
+            word, neutral_tone_with_five=True, style=Style.FINALS_TONE3
+        )
+    else:
+        orig_initials = g2pw.lazy_pinyin(word, neutral_tone_with_five=True, style=Style.INITIALS)
+        orig_finals = g2pw.lazy_pinyin(
+            word, neutral_tone_with_five=True, style=Style.FINALS_TONE3
+        )
+
+    for c, v in zip(orig_initials, orig_finals):
+        initials.append(c)
+        finals.append(v)
+    return initials, finals
+
+
+def _g2p(segments):
+    phones_list = []
+    word2ph = []
+    for seg in segments:
+        pinyins = []
+        # Replace all English words in the sentence
+        seg = re.sub("[a-zA-Z]+", "", seg)
+        seg_cut = psg.lcut(seg)
+        initials = []
+        finals = []
+        seg_cut = tone_modifier.pre_merge_for_modify(seg_cut)
+        for word, pos in seg_cut:
+            if pos == "eng":
+                continue
+            sub_initials, sub_finals = _get_initials_finals(word)
+            sub_finals = tone_modifier.modified_tone(word, pos, sub_finals)
+            initials.append(sub_initials)
+            finals.append(sub_finals)
+            # assert len(sub_initials) == len(sub_finals) == len(word)
+        initials = sum(initials, [])
+        finals = sum(finals, [])
+        #
+        for c, v in zip(initials, finals):
+            raw_pinyin = c + v
+            # NOTE: post process for pypinyin outputs
+            # we discriminate i, ii and iii
+            if c == v:
+                assert c in punctuation
+                phone = [c]
+                word2ph.append(1)
+            else:
+                v_without_tone = v[:-1]
+                tone = v[-1]
+
+                pinyin = c + v_without_tone
+                assert tone in "12345"
+
+                if c:
+                    # 多音节
+                    v_rep_map = {
+                        "uei": "ui",
+                        "iou": "iu",
+                        "uen": "un",
+                    }
+                    if v_without_tone in v_rep_map.keys():
+                        pinyin = c + v_rep_map[v_without_tone]
+                else:
+                    # 单音节
+                    pinyin_rep_map = {
+                        "ing": "ying",
+                        "i": "yi",
+                        "in": "yin",
+                        "u": "wu",
+                    }
+                    if pinyin in pinyin_rep_map.keys():
+                        pinyin = pinyin_rep_map[pinyin]
+                    else:
+                        single_rep_map = {
+                            "v": "yu",
+                            "e": "e",
+                            "i": "y",
+                            "u": "w",
+                        }
+                        if pinyin[0] in single_rep_map.keys():
+                            pinyin = single_rep_map[pinyin[0]] + pinyin[1:]
+
+                assert pinyin in pinyin_to_symbol_map.keys(), (pinyin, seg, raw_pinyin)
+                new_c, new_v = pinyin_to_symbol_map[pinyin].split(" ")
+                new_v = new_v + tone
+                phone = [new_c, new_v]
+                word2ph.append(len(phone))
+
+            phones_list += phone
+    return phones_list, word2ph
+
+
+def text_normalize(text):
+    # https://github.com/PaddlePaddle/PaddleSpeech/tree/develop/paddlespeech/t2s/frontend/zh_normalization
+    tx = TextNormalizer()
+    sentences = tx.normalize(text)
+    dest_text = ""
+    for sentence in sentences:
+        dest_text += replace_punctuation(sentence)
+    return dest_text
+
+
+if __name__ == "__main__":
+    text = "啊——但是《原神》是由,米哈\游自主，研发的一款全.新开放世界.冒险游戏"
+    text = "呣呣呣～就是…大人的鼹鼠党吧？"
+    text = "你好"
+    text = text_normalize(text)
+    print(g2p(text))
+
+
+# # 示例用法
+# text = "这是一个示例文本：,你好！这是一个测试..."
+# print(g2p_paddle(text))  # 输出: 这是一个示例文本你好这是一个测试

GPT_SoVITS/text/cleaner.py

@@ -1,6 +1,13 @@
-from text import chinese, japanese, cleaned_text_to_sequence, symbols, english
+from text import japanese, cleaned_text_to_sequence, english,korean,cantonese
+import os
+if os.environ.get("version","v1")=="v1":
+    from text import chinese
+    from text.symbols import symbols
+else:
+    from text import chinese2 as chinese
+    from text.symbols2 import symbols
 
-language_module_map = {"zh": chinese, "ja": japanese, "en": english}
+language_module_map = {"zh": chinese, "ja": japanese, "en": english, "ko": korean,"yue":cantonese}
 special = [
     # ("%", "zh", "SP"),
     ("￥", "zh", "SP2"),
@@ -17,8 +24,11 @@ def clean_text(text, language):
         if special_s in text and language == special_l:
             return clean_special(text, language, special_s, target_symbol)
     language_module = language_module_map[language]
-    norm_text = language_module.text_normalize(text)
+    if hasattr(language_module,"text_normalize"):
-    if language == "zh":
+        norm_text = language_module.text_normalize(text)
+    else:
+        norm_text=text
+    if language == "zh" or language=="yue":##########
         phones, word2ph = language_module.g2p(norm_text)
         assert len(phones) == sum(word2ph)
         assert len(norm_text) == len(word2ph)

GPT_SoVITS/text/korean.py

@@ -0,0 +1,262 @@
+import re
+from jamo import h2j, j2hcj
+import ko_pron
+from g2pk2 import G2p
+from text import symbols
+
+# This is a list of Korean classifiers preceded by pure Korean numerals.
+_korean_classifiers = '군데 권 개 그루 닢 대 두 마리 모 모금 뭇 발 발짝 방 번 벌 보루 살 수 술 시 쌈 움큼 정 짝 채 척 첩 축 켤레 톨 통'
+
+# List of (hangul, hangul divided) pairs:
+_hangul_divided = [(re.compile('%s' % x[0]), x[1]) for x in [
+    # ('ㄳ', 'ㄱㅅ'),   # g2pk2, A Syllable-ending Rule
+    # ('ㄵ', 'ㄴㅈ'),
+    # ('ㄶ', 'ㄴㅎ'),
+    # ('ㄺ', 'ㄹㄱ'),
+    # ('ㄻ', 'ㄹㅁ'),
+    # ('ㄼ', 'ㄹㅂ'),
+    # ('ㄽ', 'ㄹㅅ'),
+    # ('ㄾ', 'ㄹㅌ'),
+    # ('ㄿ', 'ㄹㅍ'),
+    # ('ㅀ', 'ㄹㅎ'),
+    # ('ㅄ', 'ㅂㅅ'),
+    ('ㅘ', 'ㅗㅏ'),
+    ('ㅙ', 'ㅗㅐ'),
+    ('ㅚ', 'ㅗㅣ'),
+    ('ㅝ', 'ㅜㅓ'),
+    ('ㅞ', 'ㅜㅔ'),
+    ('ㅟ', 'ㅜㅣ'),
+    ('ㅢ', 'ㅡㅣ'),
+    ('ㅑ', 'ㅣㅏ'),
+    ('ㅒ', 'ㅣㅐ'),
+    ('ㅕ', 'ㅣㅓ'),
+    ('ㅖ', 'ㅣㅔ'),
+    ('ㅛ', 'ㅣㅗ'),
+    ('ㅠ', 'ㅣㅜ')
+]]
+
+# List of (Latin alphabet, hangul) pairs:
+_latin_to_hangul = [(re.compile('%s' % x[0], re.IGNORECASE), x[1]) for x in [
+    ('a', '에이'),
+    ('b', '비'),
+    ('c', '시'),
+    ('d', '디'),
+    ('e', '이'),
+    ('f', '에프'),
+    ('g', '지'),
+    ('h', '에이치'),
+    ('i', '아이'),
+    ('j', '제이'),
+    ('k', '케이'),
+    ('l', '엘'),
+    ('m', '엠'),
+    ('n', '엔'),
+    ('o', '오'),
+    ('p', '피'),
+    ('q', '큐'),
+    ('r', '아르'),
+    ('s', '에스'),
+    ('t', '티'),
+    ('u', '유'),
+    ('v', '브이'),
+    ('w', '더블유'),
+    ('x', '엑스'),
+    ('y', '와이'),
+    ('z', '제트')
+]]
+
+# List of (ipa, lazy ipa) pairs:
+_ipa_to_lazy_ipa = [(re.compile('%s' % x[0], re.IGNORECASE), x[1]) for x in [
+    ('t͡ɕ','ʧ'),
+    ('d͡ʑ','ʥ'),
+    ('ɲ','n^'),
+    ('ɕ','ʃ'),
+    ('ʷ','w'),
+    ('ɭ','l`'),
+    ('ʎ','ɾ'),
+    ('ɣ','ŋ'),
+    ('ɰ','ɯ'),
+    ('ʝ','j'),
+    ('ʌ','ə'),
+    ('ɡ','g'),
+    ('\u031a','#'),
+    ('\u0348','='),
+    ('\u031e',''),
+    ('\u0320',''),
+    ('\u0339','')
+]]
+
+
+def fix_g2pk2_error(text):
+    new_text = ""
+    i = 0
+    while i < len(text) - 4:
+        if (text[i:i+3] == 'ㅇㅡㄹ' or text[i:i+3] == 'ㄹㅡㄹ') and text[i+3] == ' ' and text[i+4] == 'ㄹ':
+            new_text += text[i:i+3] + ' ' + 'ㄴ'
+            i += 5
+        else:
+            new_text += text[i]
+            i += 1
+
+    new_text += text[i:]
+    return new_text
+
+
+def latin_to_hangul(text):
+    for regex, replacement in _latin_to_hangul:
+        text = re.sub(regex, replacement, text)
+    return text
+
+
+def divide_hangul(text):
+    text = j2hcj(h2j(text))
+    for regex, replacement in _hangul_divided:
+        text = re.sub(regex, replacement, text)
+    return text
+
+
+def hangul_number(num, sino=True):
+    '''Reference https://github.com/Kyubyong/g2pK'''
+    num = re.sub(',', '', num)
+
+    if num == '0':
+        return '영'
+    if not sino and num == '20':
+        return '스무'
+
+    digits = '123456789'
+    names = '일이삼사오육칠팔구'
+    digit2name = {d: n for d, n in zip(digits, names)}
+
+    modifiers = '한 두 세 네 다섯 여섯 일곱 여덟 아홉'
+    decimals = '열 스물 서른 마흔 쉰 예순 일흔 여든 아흔'
+    digit2mod = {d: mod for d, mod in zip(digits, modifiers.split())}
+    digit2dec = {d: dec for d, dec in zip(digits, decimals.split())}
+
+    spelledout = []
+    for i, digit in enumerate(num):
+        i = len(num) - i - 1
+        if sino:
+            if i == 0:
+                name = digit2name.get(digit, '')
+            elif i == 1:
+                name = digit2name.get(digit, '') + '십'
+                name = name.replace('일십', '십')
+        else:
+            if i == 0:
+                name = digit2mod.get(digit, '')
+            elif i == 1:
+                name = digit2dec.get(digit, '')
+        if digit == '0':
+            if i % 4 == 0:
+                last_three = spelledout[-min(3, len(spelledout)):]
+                if ''.join(last_three) == '':
+                    spelledout.append('')
+                    continue
+            else:
+                spelledout.append('')
+                continue
+        if i == 2:
+            name = digit2name.get(digit, '') + '백'
+            name = name.replace('일백', '백')
+        elif i == 3:
+            name = digit2name.get(digit, '') + '천'
+            name = name.replace('일천', '천')
+        elif i == 4:
+            name = digit2name.get(digit, '') + '만'
+            name = name.replace('일만', '만')
+        elif i == 5:
+            name = digit2name.get(digit, '') + '십'
+            name = name.replace('일십', '십')
+        elif i == 6:
+            name = digit2name.get(digit, '') + '백'
+            name = name.replace('일백', '백')
+        elif i == 7:
+            name = digit2name.get(digit, '') + '천'
+            name = name.replace('일천', '천')
+        elif i == 8:
+            name = digit2name.get(digit, '') + '억'
+        elif i == 9:
+            name = digit2name.get(digit, '') + '십'
+        elif i == 10:
+            name = digit2name.get(digit, '') + '백'
+        elif i == 11:
+            name = digit2name.get(digit, '') + '천'
+        elif i == 12:
+            name = digit2name.get(digit, '') + '조'
+        elif i == 13:
+            name = digit2name.get(digit, '') + '십'
+        elif i == 14:
+            name = digit2name.get(digit, '') + '백'
+        elif i == 15:
+            name = digit2name.get(digit, '') + '천'
+        spelledout.append(name)
+    return ''.join(elem for elem in spelledout)
+
+
+def number_to_hangul(text):
+    '''Reference https://github.com/Kyubyong/g2pK'''
+    tokens = set(re.findall(r'(\d[\d,]*)([\uac00-\ud71f]+)', text))
+    for token in tokens:
+        num, classifier = token
+        if classifier[:2] in _korean_classifiers or classifier[0] in _korean_classifiers:
+            spelledout = hangul_number(num, sino=False)
+        else:
+            spelledout = hangul_number(num, sino=True)
+        text = text.replace(f'{num}{classifier}', f'{spelledout}{classifier}')
+    # digit by digit for remaining digits
+    digits = '0123456789'
+    names = '영일이삼사오육칠팔구'
+    for d, n in zip(digits, names):
+        text = text.replace(d, n)
+    return text
+
+
+def korean_to_lazy_ipa(text):
+    text = latin_to_hangul(text)
+    text = number_to_hangul(text)
+    text=re.sub('[\uac00-\ud7af]+',lambda x:ko_pron.romanise(x.group(0),'ipa').split('] ~ [')[0],text)
+    for regex, replacement in _ipa_to_lazy_ipa:
+        text = re.sub(regex, replacement, text)
+    return text
+
+_g2p=G2p()
+def korean_to_ipa(text):
+    text = latin_to_hangul(text)
+    text = number_to_hangul(text)
+    text = _g2p(text)
+    text = fix_g2pk2_error(text)
+    text = korean_to_lazy_ipa(text)
+    return text.replace('ʧ','tʃ').replace('ʥ','dʑ')
+
+def post_replace_ph(ph):
+    rep_map = {
+        "：": ",",
+        "；": ",",
+        "，": ",",
+        "。": ".",
+        "！": "!",
+        "？": "?",
+        "\n": ".",
+        "·": ",",
+        "、": ",",
+        "...": "…",
+        " ": "空",
+    }
+    if ph in rep_map.keys():
+        ph = rep_map[ph]
+    if ph in symbols:
+        return ph
+    if ph not in symbols:
+        ph = "停"
+    return ph
+
+def g2p(text):
+    text = latin_to_hangul(text)
+    text = _g2p(text)
+    text = divide_hangul(text)
+    text = fix_g2pk2_error(text)
+    text = re.sub(r'([\u3131-\u3163])$', r'\1.', text)
+    # text = "".join([post_replace_ph(i) for i in text])
+    text = [post_replace_ph(i) for i in text]
+    return text

GPT_SoVITS/text/symbols2.py

@@ -0,0 +1,419 @@
+import os
+
+# punctuation = ['!', '?', '…', ",", ".","@"]#@是SP停顿
+punctuation = ["!", "?", "…", ",", "."]  # @是SP停顿
+punctuation.append("-")
+pu_symbols = punctuation + ["SP", "SP2", "SP3", "UNK"]
+# pu_symbols = punctuation + ["SP", 'SP2', 'SP3','SP4', "UNK"]
+pad = "_"
+
+c = [
+    "AA",
+    "EE",
+    "OO",
+    "b",
+    "c",
+    "ch",
+    "d",
+    "f",
+    "g",
+    "h",
+    "j",
+    "k",
+    "l",
+    "m",
+    "n",
+    "p",
+    "q",
+    "r",
+    "s",
+    "sh",
+    "t",
+    "w",
+    "x",
+    "y",
+    "z",
+    "zh",
+]
+v = [
+    "E1",
+    "En1",
+    "a1",
+    "ai1",
+    "an1",
+    "ang1",
+    "ao1",
+    "e1",
+    "ei1",
+    "en1",
+    "eng1",
+    "er1",
+    "i1",
+    "i01",
+    "ia1",
+    "ian1",
+    "iang1",
+    "iao1",
+    "ie1",
+    "in1",
+    "ing1",
+    "iong1",
+    "ir1",
+    "iu1",
+    "o1",
+    "ong1",
+    "ou1",
+    "u1",
+    "ua1",
+    "uai1",
+    "uan1",
+    "uang1",
+    "ui1",
+    "un1",
+    "uo1",
+    "v1",
+    "van1",
+    "ve1",
+    "vn1",
+    "E2",
+    "En2",
+    "a2",
+    "ai2",
+    "an2",
+    "ang2",
+    "ao2",
+    "e2",
+    "ei2",
+    "en2",
+    "eng2",
+    "er2",
+    "i2",
+    "i02",
+    "ia2",
+    "ian2",
+    "iang2",
+    "iao2",
+    "ie2",
+    "in2",
+    "ing2",
+    "iong2",
+    "ir2",
+    "iu2",
+    "o2",
+    "ong2",
+    "ou2",
+    "u2",
+    "ua2",
+    "uai2",
+    "uan2",
+    "uang2",
+    "ui2",
+    "un2",
+    "uo2",
+    "v2",
+    "van2",
+    "ve2",
+    "vn2",
+    "E3",
+    "En3",
+    "a3",
+    "ai3",
+    "an3",
+    "ang3",
+    "ao3",
+    "e3",
+    "ei3",
+    "en3",
+    "eng3",
+    "er3",
+    "i3",
+    "i03",
+    "ia3",
+    "ian3",
+    "iang3",
+    "iao3",
+    "ie3",
+    "in3",
+    "ing3",
+    "iong3",
+    "ir3",
+    "iu3",
+    "o3",
+    "ong3",
+    "ou3",
+    "u3",
+    "ua3",
+    "uai3",
+    "uan3",
+    "uang3",
+    "ui3",
+    "un3",
+    "uo3",
+    "v3",
+    "van3",
+    "ve3",
+    "vn3",
+    "E4",
+    "En4",
+    "a4",
+    "ai4",
+    "an4",
+    "ang4",
+    "ao4",
+    "e4",
+    "ei4",
+    "en4",
+    "eng4",
+    "er4",
+    "i4",
+    "i04",
+    "ia4",
+    "ian4",
+    "iang4",
+    "iao4",
+    "ie4",
+    "in4",
+    "ing4",
+    "iong4",
+    "ir4",
+    "iu4",
+    "o4",
+    "ong4",
+    "ou4",
+    "u4",
+    "ua4",
+    "uai4",
+    "uan4",
+    "uang4",
+    "ui4",
+    "un4",
+    "uo4",
+    "v4",
+    "van4",
+    "ve4",
+    "vn4",
+    "E5",
+    "En5",
+    "a5",
+    "ai5",
+    "an5",
+    "ang5",
+    "ao5",
+    "e5",
+    "ei5",
+    "en5",
+    "eng5",
+    "er5",
+    "i5",
+    "i05",
+    "ia5",
+    "ian5",
+    "iang5",
+    "iao5",
+    "ie5",
+    "in5",
+    "ing5",
+    "iong5",
+    "ir5",
+    "iu5",
+    "o5",
+    "ong5",
+    "ou5",
+    "u5",
+    "ua5",
+    "uai5",
+    "uan5",
+    "uang5",
+    "ui5",
+    "un5",
+    "uo5",
+    "v5",
+    "van5",
+    "ve5",
+    "vn5",
+]
+
+v_without_tone = [
+    "E",
+    "En",
+    "a",
+    "ai",
+    "an",
+    "ang",
+    "ao",
+    "e",
+    "ei",
+    "en",
+    "eng",
+    "er",
+    "i",
+    "i0",
+    "ia",
+    "ian",
+    "iang",
+    "iao",
+    "ie",
+    "in",
+    "ing",
+    "iong",
+    "ir",
+    "iu",
+    "o",
+    "ong",
+    "ou",
+    "u",
+    "ua",
+    "uai",
+    "uan",
+    "uang",
+    "ui",
+    "un",
+    "uo",
+    "v",
+    "van",
+    "ve",
+    "vn",
+]
+
+# japanese
+ja_symbols = [
+    "I",
+    "N",
+    "U",
+    "a",
+    "b",
+    "by",
+    "ch",
+    "cl",
+    "d",
+    "dy",
+    "e",
+    "f",
+    "g",
+    "gy",
+    "h",
+    "hy",
+    "i",
+    "j",
+    "k",
+    "ky",
+    "m",
+    "my",
+    "n",
+    "ny",
+    "o",
+    "p",
+    "py",
+    "r",
+    "ry",
+    "s",
+    "sh",
+    "t",
+    "ts",
+    "u",
+    "v",
+    "w",
+    "y",
+    "z",
+    ###楼下2个留到后面加
+    # "[", #上升调型
+    # "]", #下降调型
+    # "$", #结束符
+    # "^", #开始符
+]
+
+arpa = {
+    "AH0",
+    "S",
+    "AH1",
+    "EY2",
+    "AE2",
+    "EH0",
+    "OW2",
+    "UH0",
+    "NG",
+    "B",
+    "G",
+    "AY0",
+    "M",
+    "AA0",
+    "F",
+    "AO0",
+    "ER2",
+    "UH1",
+    "IY1",
+    "AH2",
+    "DH",
+    "IY0",
+    "EY1",
+    "IH0",
+    "K",
+    "N",
+    "W",
+    "IY2",
+    "T",
+    "AA1",
+    "ER1",
+    "EH2",
+    "OY0",
+    "UH2",
+    "UW1",
+    "Z",
+    "AW2",
+    "AW1",
+    "V",
+    "UW2",
+    "AA2",
+    "ER",
+    "AW0",
+    "UW0",
+    "R",
+    "OW1",
+    "EH1",
+    "ZH",
+    "AE0",
+    "IH2",
+    "IH",
+    "Y",
+    "JH",
+    "P",
+    "AY1",
+    "EY0",
+    "OY2",
+    "TH",
+    "HH",
+    "D",
+    "ER0",
+    "CH",
+    "AO1",
+    "AE1",
+    "AO2",
+    "OY1",
+    "AY2",
+    "IH1",
+    "OW0",
+    "L",
+    "SH",
+}
+
+ko_symbols='ㄱㄴㄷㄹㅁㅂㅅㅇㅈㅊㅋㅌㅍㅎㄲㄸㅃㅆㅉㅏㅓㅗㅜㅡㅣㅐㅔ空停'
+# ko_symbols='ㄱㄴㄷㄹㅁㅂㅅㅇㅈㅊㅋㅌㅍㅎㄲㄸㅃㅆㅉㅏㅓㅗㅜㅡㅣㅐㅔ '
+
+yue_symbols={'Yeot3', 'Yip1', 'Yyu3', 'Yeng4', 'Yut5', 'Yaan5', 'Ym5', 'Yaan6', 'Yang1', 'Yun4', 'Yon2', 'Yui5', 'Yun2', 'Yat3', 'Ye', 'Yeot1', 'Yoeng5', 'Yoek2', 'Yam2', 'Yeon6', 'Yu6', 'Yiu3', 'Yaang6', 'Yp5', 'Yai4', 'Yoek4', 'Yit6', 'Yam5', 'Yoeng6', 'Yg1', 'Yk3', 'Yoe4', 'Yam3', 'Yc', 'Yyu4', 'Yyut1', 'Yiu4', 'Ying3', 'Yip3', 'Yaap3', 'Yau3', 'Yan4', 'Yau1', 'Yap4', 'Yk6', 'Yok3', 'Yai1', 'Yeot6', 'Yan2', 'Yoek6', 'Yt1', 'Yoi1', 'Yit5', 'Yn4', 'Yaau3', 'Yau4', 'Yuk6', 'Ys', 'Yuk', 'Yin6', 'Yung6', 'Ya', 'You', 'Yaai5', 'Yau5', 'Yoi3', 'Yaak3', 'Yaat3', 'Ying2', 'Yok5', 'Yeng2', 'Yyut3', 'Yam1', 'Yip5', 'You1', 'Yam6', 'Yaa5', 'Yi6', 'Yek4', 'Yyu2', 'Yuk5', 'Yaam1', 'Yang2', 'Yai', 'Yiu6', 'Yin4', 'Yok4', 'Yot3', 'Yui2', 'Yeoi5', 'Yyun6', 'Yyu5', 'Yoi5', 'Yeot2', 'Yim4', 'Yeoi2', 'Yaan1', 'Yang6', 'Yong1', 'Yaang4', 'Yung5', 'Yeon1', 'Yin2', 'Ya3', 'Yaang3', 'Yg', 'Yk2', 'Yaau5', 'Yut1', 'Yt5', 'Yip4', 'Yung4', 'Yj', 'Yong3', 'Ya1', 'Yg6', 'Yaau6', 'Yit3', 'Yun3', 'Ying1', 'Yn2', 'Yg4', 'Yl', 'Yp3', 'Yn3', 'Yak1', 'Yang5', 'Yoe6', 'You2', 'Yap2', 'Yak2', 'Yt3', 'Yot5', 'Yim2', 'Yi1', 'Yn6', 'Yaat5', 'Yaam3', 'Yoek5', 'Ye3', 'Yeon4', 'Yaa2', 'Yu3', 'Yim6', 'Ym', 'Yoe3', 'Yaai2', 'Ym2', 'Ya6', 'Yeng6', 'Yik4', 'Yot4', 'Yaai4', 'Yyun3', 'Yu1', 'Yoeng1', 'Yaap2', 'Yuk3', 'Yoek3', 'Yeng5', 'Yeoi1', 'Yiu2', 'Yok1', 'Yo1', 'Yoek1', 'Yoeng2', 'Yeon5', 'Yiu1', 'Yoeng4', 'Yuk2', 'Yat4', 'Yg5', 'Yut4', 'Yan6', 'Yin3', 'Yaa6', 'Yap1', 'Yg2', 'Yoe5', 'Yt4', 'Ya5', 'Yo4', 'Yyu1', 'Yak3', 'Yeon2', 'Yong4', 'Ym1', 'Ye2', 'Yaang5', 'Yoi2', 'Yeng3', 'Yn', 'Yyut4', 'Yau', 'Yaak2', 'Yaan4', 'Yek2', 'Yin1', 'Yi5', 'Yoe2', 'Yei5', 'Yaat6', 'Yak5', 'Yp6', 'Yok6', 'Yei2', 'Yaap1', 'Yyut5', 'Yi4', 'Yim1', 'Yk5', 'Ye4', 'Yok2', 'Yaam6', 'Yat2', 'Yon6', 'Yei3', 'Yyu6', 'Yeot5', 'Yk4', 'Yai6', 'Yd', 'Yg3', 'Yei6', 'Yau2', 'Yok', 'Yau6', 'Yung3', 'Yim5', 'Yut6', 'Yit1', 'Yon3', 'Yat1', 'Yaam2', 'Yyut2', 'Yui6', 'Yt2', 'Yek6', 'Yt', 'Ye6', 'Yang3', 'Ying6', 'Yaau1', 'Yeon3', 'Yng', 'Yh', 'Yang4', 'Ying5', 'Yaap6', 'Yoeng3', 'Yyun4', 'You3', 'Yan5', 'Yat5', 'Yot1', 'Yun1', 'Yi3', 'Yaa1', 'Yaap4', 'You6', 'Yaang2', 'Yaap5', 'Yaa3', 'Yaak6', 'Yeng1', 'Yaak1', 'Yo5', 'Yoi4', 'Yam4', 'Yik1', 'Ye1', 'Yai5', 'Yung1', 'Yp2', 'Yui4', 'Yaak4', 'Yung2', 'Yak4', 'Yaat4', 'Yeoi4', 'Yut2', 'Yin5', 'Yaau4', 'Yap6', 'Yb', 'Yaam4', 'Yw', 'Yut3', 'Yong2', 'Yt6', 'Yaai6', 'Yap5', 'Yik5', 'Yun6', 'Yaam5', 'Yun5', 'Yik3', 'Ya2', 'Yyut6', 'Yon4', 'Yk1', 'Yit4', 'Yak6', 'Yaan2', 'Yuk1', 'Yai2', 'Yik2', 'Yaat2', 'Yo3', 'Ykw', 'Yn5', 'Yaa', 'Ye5', 'Yu4', 'Yei1', 'Yai3', 'Yyun5', 'Yip2', 'Yaau2', 'Yiu5', 'Ym4', 'Yeoi6', 'Yk', 'Ym6', 'Yoe1', 'Yeoi3', 'Yon', 'Yuk4', 'Yaai3', 'Yaa4', 'Yot6', 'Yaang1', 'Yei4', 'Yek1', 'Yo', 'Yp', 'Yo6', 'Yp4', 'Yan3', 'Yoi', 'Yap3', 'Yek3', 'Yim3', 'Yz', 'Yot2', 'Yoi6', 'Yit2', 'Yu5', 'Yaan3', 'Yan1', 'Yon5', 'Yp1', 'Yong5', 'Ygw', 'Yak', 'Yat6', 'Ying4', 'Yu2', 'Yf', 'Ya4', 'Yon1', 'You4', 'Yik6', 'Yui1', 'Yaat1', 'Yeot4', 'Yi2', 'Yaai1', 'Yek5', 'Ym3', 'Yong6', 'You5', 'Yyun1', 'Yn1', 'Yo2', 'Yip6', 'Yui3', 'Yaak5', 'Yyun2'}
+
+# symbols = [pad] + c + v + ja_symbols + pu_symbols + list(arpa)+list(ko_symbols)#+list(yue_symbols)###直接这么加yue顺序乱了
+symbols = [pad] + c + v + ja_symbols + pu_symbols + list(arpa)
+symbols = sorted(set(symbols))
+# print(len(symbols))
+symbols+=["[","]"]##日文新增上升下降调型
+symbols+=sorted(list(ko_symbols))
+symbols+=sorted(list(yue_symbols))##新加的yue统一摆在后头#已查过开头加Y后没有重复，韩文显然不会重复
+# print(len(symbols))
+if __name__ == "__main__":
+    print(len(symbols))
+'''
+粤语：
+    732-353=379
+韩文+粤语：
+    732-322=410
+'''