WeTextProcessing 1.0.2__py3-none-any.whl → 1.0.3__py3-none-any.whl

{WeTextProcessing-1.0.2.dist-info → WeTextProcessing-1.0.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: WeTextProcessing
-Version: 1.0.2
+Version: 1.0.3
 Summary: WeTextProcessing, including TN & ITN
 Home-page: https://github.com/wenet-e2e/WeTextProcessing
 Author: Zhendong Peng, Xingchen Song
@@ -51,14 +51,34 @@ weitn --text "二点五平方电线"
 Python usage:
 
 ```py
-# tn usage
->>> from tn.chinese.normalizer import Normalizer
->>> normalizer = Normalizer()
->>> normalizer.normalize("2.5平方电线")
-# itn usage
->>> from itn.chinese.inverse_normalizer import InverseNormalizer
->>> invnormalizer = InverseNormalizer()
->>> invnormalizer.normalize("二点五平方电线")
+from itn.chinese.inverse_normalizer import InverseNormalizer
+from tn.chinese.normalizer import Normalizer as ZhNormalizer
+from tn.english.normalizer import Normalizer as EnNormalizer
+
+# NOTE(xcsong): 和默认参数不一致时，必须重新构图，要重新构图请务必指定 `overwrite_cache=True`
+#               When the parameters differ from the defaults, it is mandatory to re-compose. To re-compose, please ensure you specify `overwrite_cache=True`.
+
+zh_tn_text = "你好 WeTextProcessing 1.0，船新版本儿，船新体验儿，简直666，9和10"
+zh_itn_text = "你好 WeTextProcessing 一点零，船新版本儿，船新体验儿，简直六六六，九和六"
+en_tn_text = "Hello WeTextProcessing 1.0, life is short, just use wetext, 666, 9 and 10"
+zh_tn_model = ZhNormalizer(remove_erhua=True, overwrite_cache=True)
+zh_itn_model = InverseNormalizer(enable_0_to_9=False, overwrite_cache=True)
+en_tn_model = EnNormalizer(overwrite_cache=True)
+print("中文 TN (去除儿化音，重新在线构图):\n\t{} => {}".format(zh_tn_text, zh_tn_model.normalize(zh_tn_text)))
+print("中文ITN (小于10的单独数字不转换，重新在线构图):\n\t{} => {}".format(zh_itn_text, zh_itn_model.normalize(zh_itn_text)))
+print("英文 TN (暂时还没有可控的选项，后面会加...):\n\t{} => {}\n".format(en_tn_text, en_tn_model.normalize(en_tn_text)))
+
+zh_tn_model = ZhNormalizer(overwrite_cache=False)
+zh_itn_model = InverseNormalizer(overwrite_cache=False)
+en_tn_model = EnNormalizer(overwrite_cache=False)
+print("中文 TN (复用之前编译好的图):\n\t{} => {}".format(zh_tn_text, zh_tn_model.normalize(zh_tn_text)))
+print("中文ITN (复用之前编译好的图):\n\t{} => {}".format(zh_itn_text, zh_itn_model.normalize(zh_itn_text)))
+print("英文 TN (复用之前编译好的图):\n\t{} => {}\n".format(en_tn_text, en_tn_model.normalize(en_tn_text)))
+
+zh_tn_model = ZhNormalizer(remove_erhua=False, overwrite_cache=True)
+zh_itn_model = InverseNormalizer(enable_0_to_9=True, overwrite_cache=True)
+print("中文 TN (不去除儿化音，重新在线构图):\n\t{} => {}".format(zh_tn_text, zh_tn_model.normalize(zh_tn_text)))
+print("中文ITN (小于10的单独数字也进行转换，重新在线构图):\n\t{} => {}\n".format(zh_itn_text, zh_itn_model.normalize(zh_itn_text)))
 ```
 
 #### 1.2 Advanced Usage:

{WeTextProcessing-1.0.2.dist-info → WeTextProcessing-1.0.3.dist-info}/RECORD

@@ -1,14 +1,14 @@
 itn/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 itn/__main__.py,sha256=jFh8bEB_lC5y7ab9VMqAxWRaI8D-SpBSTiHfMJC1b10,65
 itn/main.py,sha256=NnT9Sh3VTEI0sX_DO9IGVsXCWtZyfRbzj9Sj_6cBxqM,2610
-itn/zh_itn_tagger.fst,sha256=32FSrQ6gcFYNTUYaHinnrdaMZUpAwuDmvu-EPnki8aI,1172366
+itn/zh_itn_tagger.fst,sha256=WUG3WqZ7ZbyqbbPvueFM-GEco96rsGNjx5ZDRg7USA8,1175402
 itn/zh_itn_verbalizer.fst,sha256=pHXRUfzls2UmWwTmHHhYmBWMusIfWpU3I7IwxeRjn28,119018
 itn/chinese/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 itn/chinese/inverse_normalizer.py,sha256=9GpVP7_BCBoCIpoZkPa8aq-n7M56GhY9l4mp3jUGwpM,3563
 itn/chinese/data/date/dd.tsv,sha256=4KnAC3ETYcPwLtiyJjJPSyU0F0UQnvTNEu3kVAG1oII,806
 itn/chinese/data/date/mm.tsv,sha256=lkWAizxrCwko42QGsLhZZUmjJ0FJw8Faii7xqHLkZec,126
 itn/chinese/data/default/blacklist.tsv,sha256=kLgMYyQjmJvc-O-2nhOF2EfBxJgNmOjLhDtbKzHenDk,8
-itn/chinese/data/default/whitelist.tsv,sha256=zPhRTIH_vxjBqLiXnDX4_9dT4pNfb0grEmi56CNjeJQ,1338
+itn/chinese/data/default/whitelist.tsv,sha256=WLnLWYcAYWQuZfEZAVJlGyDQi4C1oFXeXTa3be8kk-M,1686
 itn/chinese/data/license_plate/province.tsv,sha256=8FGMiiTQgp-PWOG9SGzLJSyIgqlB2ZWsHXIKA4CheTw,124
 itn/chinese/data/math/operator.tsv,sha256=_qGhDdskEaXRL4oKWy1lbeUWIG-ySpot9n23iu5sX9Q,47
 itn/chinese/data/measure/units_en.tsv,sha256=hjkxNHc9g6EEnrbzQy32FqNFT8fkA9MAT6VBeeZ4TW0,1087
@@ -16,6 +16,7 @@ itn/chinese/data/measure/units_zh.tsv,sha256=4X-3EKTKOiR2vYZQrHWqBAZtogpG13eO4PO
 itn/chinese/data/money/code.tsv,sha256=QXzDJD0QJm1FtsTtMIzI641ZSYReJfr7Q5ZRZYKbImM,2374
 itn/chinese/data/money/symbol.tsv,sha256=5QxiR9VnB6pX2l3EKmbdbwLLGCJrDHEpym2DxqVXRxQ,1266
 itn/chinese/data/number/digit.tsv,sha256=l-5ompNSHfBLhgWaTr-QqBh7iLN3PGty571UandT2Q4,126
+itn/chinese/data/number/digit_zh.tsv,sha256=5MZabXiLvomGd0dMjHEjppbKnnIYOATZJvs9W9iQi3E,40
 itn/chinese/data/number/dot.tsv,sha256=1xj5VPX4Qu0JiPycy8cTAqpsGAJJrbQwqab_TctCXqs,12
 itn/chinese/data/number/sign.tsv,sha256=VnjOk140OULqin_tHv_nK15DJSTjvCO-E2hYnYYSp24,31
 itn/chinese/data/number/special_dash.tsv,sha256=9zJdqG_rFu48lmeCxGLsEFOSoqUUjNVUNRtbHQ9L7Nk,661
@@ -32,7 +33,7 @@ itn/chinese/rules/date.py,sha256=8pB_S9BFKYd5rXGxzfjAcas-ecLM4dUnhNIsp_vZgI0,244
 itn/chinese/rules/fraction.py,sha256=jGhKx91TtscqPFwDtqaI0zeMCAsCFA68KPCWBXtT030,2030
 itn/chinese/rules/license_plate.py,sha256=z0bSp1RjQB_8p2FeKSgUMhYJ1CxO_3fP8pzcb75DOec,1567
 itn/chinese/rules/math.py,sha256=A_iEaTmBg95M73aIPz_L7cGLe8ZMD2ODP33a6_esfoM,1263
-itn/chinese/rules/measure.py,sha256=pPuFbFpyR9tzViEWo0XuM9_Wm4bEUgptWRP2bM918tA,3109
+itn/chinese/rules/measure.py,sha256=0ixwJ54RIJlCG0-dE4bOA9usHFRdHT7JHL6K2Bch9bs,4430
 itn/chinese/rules/money.py,sha256=fqS28BjAdGY-jQGTh3SZbd7bA1_ylFznCaMW_0e4Gr8,2271
 itn/chinese/rules/postprocessor.py,sha256=BZ1dpyaQJwxTIzsxmMrMbv5PuLbCza-163ERJ8ZC8MY,1208
 itn/chinese/rules/time.py,sha256=m99PvGEnPZSXHiu3vY8S7PzzjMdypn61u2uz6hSrUM4,2033
@@ -42,9 +43,9 @@ itn/chinese/test/normalizer_test.py,sha256=a9nqt-is07zYOl3XDAhjUyNzBlwBe3qdaHtvT
 itn/chinese/test/utils.py,sha256=aCZH5BL0dghGjemhVwfISBUxLFMG_ceMzwzuOb5ht28,1094
 tn/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 tn/__main__.py,sha256=OUMsgyp8SWb7raJY_py61uM29j023yE5po95O9yinzk,64
-tn/en_tn_tagger.fst,sha256=gqjt_9TO5osM29IWhLzHwcu-3z-NJHqkrCs04qfmANg,5418482
-tn/en_tn_verbalizer.fst,sha256=54G4Z7G1_cIs20skJCCwq9DRPmexqExTtGYiCdnB2gw,1710934
-tn/main.py,sha256=Q45id3b2g71MPRLnaHpo8iWPc38MeyZKigNd5FinNvI,3578
+tn/en_tn_tagger.fst,sha256=yYMCyIUBnfeq35yr6M0Bn_DSSs-iKoA3uLW4IbSbk-M,5248066
+tn/en_tn_verbalizer.fst,sha256=DcK9hJao5KFpWSSVsV3BhJ1QIJfIo9OHU0ahc3YbSLQ,2121910
+tn/main.py,sha256=ZQ1ZWOa-5YoU_PNDSt2P-f3RC79WUwT0UMqm4bulCIA,3587
 tn/processor.py,sha256=CUmJ7B0OyweByf7U0yi2TYqBbEhhWTcqs03_0XhTosg,5012
 tn/token_parser.py,sha256=nsTYzug-gq2HYnTJHja_HXntoRVX3gITQAlI9z8WvWU,4487
 tn/utils.py,sha256=lJLuW2jkobgXCyNYl29zhtyb3B80CS7Pqz7zqHRVr4c,2792
@@ -110,7 +111,7 @@ tn/chinese/test/token_parser_test.py,sha256=8JcSqlvaYTZ46rZgMsfAVNmcYez6t0gonhnx
 tn/chinese/test/utils.py,sha256=aCZH5BL0dghGjemhVwfISBUxLFMG_ceMzwzuOb5ht28,1094
 tn/chinese/test/whitelist_test.py,sha256=oPB-7_g8l6zhAeH5fp7NlaRJYcDNz7VAOKwhCnebGdc,1003
 tn/english/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-tn/english/normalizer.py,sha256=YNd9PYib_3m9ExMmxd0SAq0CWpEHxpEi4S0ZpNrqyVs,3921
+tn/english/normalizer.py,sha256=91v9En8-6ishv0UPaLg1wpoLqFVoEBQUZBLtWlK_zKE,4157
 tn/english/data/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 tn/english/data/suppletive.tsv,sha256=iih47clDISe35FR-p4Usu_coR3oQy1miXd4VsBJSsyA,1843
 tn/english/data/address/__init__.py,sha256=EqFAHHjK7kJqR_FV8vBLddBJWhbVD8pREEfm8C5nSEE,623
@@ -170,7 +171,7 @@ tn/english/data/whitelist/asr_with_pc.tsv,sha256=wqeEeG4O742D2vroywzbstwmr-s6lGH
 tn/english/data/whitelist/ipa_symbols.tsv,sha256=ChsK_JiztbsqNnFYNQ0xZ9QCFUxeTqRzm8vCkydQvl8,2532
 tn/english/data/whitelist/lj_speech.tsv,sha256=-sZeHERreSBZGbLiJh0q_laYvXWG5mvMAUuXkYiHQdI,263
 tn/english/data/whitelist/symbol.tsv,sha256=rIyi-7u3X-DoBslNDaAQghNVWlD6yLAgYS2nuGKQUJ4,240
-tn/english/data/whitelist/tts.tsv,sha256=JZzUz_qoLQbqkfLsluDklCIeqBHNh2nQgAWcAea03Y0,33073
+tn/english/data/whitelist/tts.tsv,sha256=8qu9zewiYFiK-PQnVVEueF6fbTpV5D2-86iF05C3cIE,33157
 tn/english/rules/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 tn/english/rules/cardinal.py,sha256=w3HvKbq86mZjb3MhXc96D_Ajaem1raCHs0pCGvhQaRs,8424
 tn/english/rules/date.py,sha256=QoUczkPwAeYanT41_8Qpi_FqvuSfvXX5LHdEgcsLkVA,13683
@@ -180,7 +181,7 @@ tn/english/rules/fraction.py,sha256=ltGhbd1Qeb-BsuVCPYq8cc__k0Xkkn4WBUiz6dFs8J8,
 tn/english/rules/measure.py,sha256=DjQnq_9UXL-pHdqaHkp3oUv4qEhiGB_764YmUEq9v6w,14025
 tn/english/rules/money.py,sha256=0hjQWUdI4wWsOVKkkBs5lvFsQaMIWOQcs6oOLBhQzDM,5457
 tn/english/rules/ordinal.py,sha256=xFUaNvPx2rKDJz3TMiE2RKje6gcvcVImr3sXixWfOKU,3989
-tn/english/rules/punctuation.py,sha256=he6GQ-6CUf9lMzPDCU2PW_RqboQHwhb2V5M5mdBdFHo,3107
+tn/english/rules/punctuation.py,sha256=Pno-BxVZrFVzNVu8X5f8EpoxMv4JOpBuJekBhM8fT5Y,3274
 tn/english/rules/range.py,sha256=lEjLyFBJSCHnHlQBdrzhiUdNVIiwTQClPz4DNRqHnB4,5730
 tn/english/rules/roman.py,sha256=pQFuSIjzWnMN-jiSuVqv3AxJEA9kA8s0NUgGpCCVbiU,6051
 tn/english/rules/telephone.py,sha256=JkhHZU1O2h4a_W3QHVBY1lOQvq4Tys5JwAlL1Kkt8Pg,8275
@@ -204,9 +205,9 @@ tn/english/test/time_test.py,sha256=4E_PS-D5caTUZgrI-UL7q0aoqmtD8np7yeXUkGyc1Qo,
 tn/english/test/utils.py,sha256=aCZH5BL0dghGjemhVwfISBUxLFMG_ceMzwzuOb5ht28,1094
 tn/english/test/whitelist_test.py,sha256=fnlpnwiDgQWsl5ONZklWXEyLQoUSuwhdowcV7a8fXvw,1023
 tn/english/test/word_test.py,sha256=ACguloXPl4Ttw8fj9F_wvr1coaQAoj_MLOJ74wyy-WY,954
-WeTextProcessing-1.0.2.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-WeTextProcessing-1.0.2.dist-info/METADATA,sha256=vINHY2ibaBLdN56EpTFgkk5Wn-jde8RfYy1gS6qnn7A,5178
-WeTextProcessing-1.0.2.dist-info/WHEEL,sha256=GJ7t_kWBFywbagK5eo9IoUwLW6oyOeTKmQ-9iHFVNxQ,92
-WeTextProcessing-1.0.2.dist-info/entry_points.txt,sha256=iAd8tFSNA8AY5fUECPlk9KaDFC3ISvDjoKAZVsCWPjE,61
-WeTextProcessing-1.0.2.dist-info/top_level.txt,sha256=6NasE_UheQezX9_0V3MbiOSnTQpvM_funMjv4wpMGgA,7
-WeTextProcessing-1.0.2.dist-info/RECORD,,
+WeTextProcessing-1.0.3.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+WeTextProcessing-1.0.3.dist-info/METADATA,sha256=gM1M2EyE0OQASspFF5zh_kP39uSUmSsfx6znL3P0_w4,7181
+WeTextProcessing-1.0.3.dist-info/WHEEL,sha256=GJ7t_kWBFywbagK5eo9IoUwLW6oyOeTKmQ-9iHFVNxQ,92
+WeTextProcessing-1.0.3.dist-info/entry_points.txt,sha256=iAd8tFSNA8AY5fUECPlk9KaDFC3ISvDjoKAZVsCWPjE,61
+WeTextProcessing-1.0.3.dist-info/top_level.txt,sha256=6NasE_UheQezX9_0V3MbiOSnTQpvM_funMjv4wpMGgA,7
+WeTextProcessing-1.0.3.dist-info/RECORD,,

itn/chinese/data/default/whitelist.tsv

@@ -61,3 +61,18 @@
 五台山	五台山
 六盘水	六盘水
 八宿	八宿
+十二五	十二五
+十三五	十三五
+十四五	十四五
+几十万	几十万
+几百万	几百万
+几千万	几千万
+十几万	十几万
+二十几万	二十几万
+三十几万	三十几万
+四十几万	四十几万
+五十几万	五十几万
+六十几万	六十几万
+七十几万	七十几万
+八十几万	八十几万
+九十几万	九十几万

itn/chinese/data/number/digit_zh.tsv

@@ -0,0 +1,10 @@
+一
+二
+两
+三
+四
+五
+六
+七
+八
+九

itn/chinese/rules/measure.py

@@ -16,7 +16,7 @@ from itn.chinese.rules.cardinal import Cardinal
 from tn.processor import Processor
 from tn.utils import get_abs_path
 
-from pynini import string_file, accep, cross
+from pynini import string_file, accep, cross, union
 from pynini.lib.pynutil import delete, insert, add_weight
 
 
@@ -36,6 +36,11 @@ class Measure(Processor):
             get_abs_path('../itn/chinese/data/measure/units_zh.tsv'))
         sign = string_file(
             get_abs_path('../itn/chinese/data/number/sign.tsv'))  # + -
+        digit = string_file(
+            get_abs_path('../itn/chinese/data/number/digit.tsv'))  # 1 ~ 9
+        digit_zh = string_file(
+            get_abs_path('../itn/chinese/data/number/digit_zh.tsv'))  # 1 ~ 9
+        addzero = insert('0')
         to = cross('到', '~') | cross('到百分之', '~')
 
         units = add_weight(
@@ -55,8 +60,35 @@ class Measure(Processor):
 
         # 十千米每小时 => 10km/h, 十一到一百千米每小时 => 11~100km/h
         measure = number + (to + number).ques + units
-        tagger = insert('value: "') + (measure | percent) + insert('"')
 
+        # XXX: 特殊case处理, ignore enable_standalone_number
+        # digit + union("百", "千", "万") + digit + unit
+        unit_sp_case1 = [
+            '年',
+            '月',
+            '个月',
+            '周',
+            '天',
+            '位',
+            '次',
+            '个',
+            '顿',
+        ]
+        if self.enable_0_to_9:
+            measure_sp = add_weight(
+                ((digit + delete('百') + add_weight(addzero**2, 1.0)) |
+                 (digit + delete('千') + add_weight(addzero**3, 1.0)) |
+                 (digit + delete('万') + add_weight(addzero**4, 1.0))) +
+                insert(' ') + digit + union(*unit_sp_case1), -0.5)
+        else:
+            measure_sp = add_weight(
+                ((digit + delete('百') + add_weight(addzero**2, 1.0)) |
+                 (digit + delete('千') + add_weight(addzero**3, 1.0)) |
+                 (digit + delete('万') + add_weight(addzero**4, 1.0))) +
+                digit_zh + union(*unit_sp_case1), -0.5)
+
+        tagger = insert('value: "') + (measure | measure_sp
+                                       | percent) + insert('"')
         # 每小时十千米 => 10km/h, 每小时三十到三百一十一千米 => 30~311km/h
         tagger |= (insert('denominator: "') + delete('每') + units +
                    insert('" numerator: "') + measure + insert('"'))

tn/english/data/whitelist/tts.tsv

@@ -3061,3 +3061,9 @@ Z. Y.	ZY
 Z.Y.	ZY
 Z. Z.	ZZ
 Z.Z.	ZZ
+here's	here is
+you're	you are
+i'm	i am
+Here's	Here is
+You're	You are
+I'm	I am

tn/english/normalizer.py

@@ -57,15 +57,25 @@ class Normalizer(Processor):
         punct = add_weight(Punctuation().tagger, 2.00)
         rang = add_weight(Range().tagger, 1.01)
         # TODO(xcsong): add roman
-        tagger = punct.star + \
-            (cardinal | ordinal | word
-             | date | decimal | fraction
-             | time | measure | money
-             | telephone | electronic
+        tagger = \
+            (cardinal
+             | ordinal
+             | word
+             | date
+             | decimal
+             | fraction
+             | time
+             | measure
+             | money
+             | telephone
+             | electronic
              | whitelist
-             | rang).optimize() + (punct.plus | self.DELETE_SPACE)
-        # delete the last space
-        self.tagger = tagger.star @ self.build_rule(delete(' '), r='[EOS]')
+             | rang
+             | punct
+             ).optimize() + (punct.plus | self.DELETE_SPACE)
+        # delete the first and last space
+        self.tagger = (delete(' ').star + tagger.star) @ self.build_rule(
+            delete(' '), r='[EOS]')
 
     def build_verbalizer(self):
         cardinal = Cardinal().verbalizer
@@ -83,14 +93,20 @@ class Normalizer(Processor):
         punct = Punctuation().verbalizer
         rang = Range().verbalizer
         verbalizer = \
-            (cardinal | ordinal | word
-             | date | decimal
-             | fraction | time
-             | measure | money
+            (cardinal
+             | ordinal
+             | word
+             | date
+             | decimal
+             | fraction
+             | time
+             | measure
+             | money
              | telephone
              | electronic
              | whitelist
              | punct
-             | rang).optimize() + (punct.plus | self.INSERT_SPACE)
+             | rang
+             ).optimize() + (punct.plus | self.INSERT_SPACE)
         self.verbalizer = verbalizer.star @ self.build_rule(delete(' '),
                                                             r='[EOS]')

tn/english/rules/punctuation.py

@@ -61,14 +61,15 @@ class Punctuation(Processor):
         punct = closure(self.punct | cross('\\', '\\\\\\') | cross('"', '\\"'),
                         1)
 
-        emphasis = (
+        self.emphasis = (
             accep("<") +
             ((
                 closure(self.NOT_SPACE - union("<", ">"), 1) +  # noqa
                 closure(accep("/"), 0, 1))  # noqa
              | (accep("/") + closure(self.NOT_SPACE - union("<", ">"), 1))) +
             accep(">"))  # noqa
-        punct = plurals._priority_union(emphasis, punct, closure(self.VCHAR))
+        punct = plurals._priority_union(self.emphasis, punct,
+                                        closure(self.VCHAR))
 
         self.graph = punct
         final_graph = insert("v: \"") + add_weight(
@@ -78,7 +79,10 @@ class Punctuation(Processor):
 
     def build_verbalizer(self):
         punct = closure(
-            self.punct | cross('\\\\\\', '\\') | cross('\\"', '"')
-            | accep(" "), 1)
-        verbalizer = delete('v: "') + punct + delete('"')
+            self.punct | self.emphasis | cross('\\\\\\', '\\')
+            | cross('\\"', '"'), 1)
+        verbalizer = delete('v: "') + add_weight(accep(" "), -1.0).star \
+            + punct \
+            + add_weight(accep(" "), -1.0).star \
+            + delete('"')
         self.verbalizer = self.delete_tokens(verbalizer)

tn/main.py

@@ -73,9 +73,8 @@ def main():
             full_to_half=str2bool(args.full_to_half),
             tag_oov=str2bool(args.tag_oov))
     elif args.language == "en":
-        normalizer = EnNormalizer(
-            cache_dir=args.cache_dir,
-            overwrite_cache=args.overwrite_cache)
+        normalizer = EnNormalizer(cache_dir=args.cache_dir,
+                                  overwrite_cache=args.overwrite_cache)
 
     if args.text:
         print(normalizer.tag(args.text))