Spaces:
Running
on
Zero
Running
on
Zero
File size: 3,560 Bytes
b96e750 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 |
# Copyright (c) 2024 Amphion.
#
# This source code is licensed under the MIT license found in the
# LICENSE file in the root directory of this source tree.
import re
"""
Text clean time
"""
# List of (regular expression, replacement) pairs for abbreviations in french:
_abbreviations = [
(re.compile("\\b%s\\." % x[0], re.IGNORECASE), x[1])
for x in [
("M", "monsieur"),
("Mlle", "mademoiselle"),
("Mlles", "mesdemoiselles"),
("Mme", "Madame"),
("Mmes", "Mesdames"),
("N.B", "nota bene"),
("M", "monsieur"),
("p.c.q", "parce que"),
("Pr", "professeur"),
("qqch", "quelque chose"),
("rdv", "rendez-vous"),
("max", "maximum"),
("min", "minimum"),
("no", "numéro"),
("adr", "adresse"),
("dr", "docteur"),
("st", "saint"),
("co", "companie"),
("jr", "junior"),
("sgt", "sergent"),
("capt", "capitain"),
("col", "colonel"),
("av", "avenue"),
("av. J.-C", "avant Jésus-Christ"),
("apr. J.-C", "après Jésus-Christ"),
("art", "article"),
("boul", "boulevard"),
("c.-à-d", "c’est-à-dire"),
("etc", "et cetera"),
("ex", "exemple"),
("excl", "exclusivement"),
("boul", "boulevard"),
]
] + [
(re.compile("\\b%s" % x[0]), x[1])
for x in [
("Mlle", "mademoiselle"),
("Mlles", "mesdemoiselles"),
("Mme", "Madame"),
("Mmes", "Mesdames"),
]
]
rep_map = {
"：": ",",
"；": ",",
"，": ",",
"。": ".",
"！": "!",
"？": "?",
"\n": ".",
"·": ",",
"、": ",",
"...": ".",
"…": ".",
"$": ".",
"“": "",
"”": "",
"‘": "",
"’": "",
"（": "",
"）": "",
"(": "",
")": "",
"《": "",
"》": "",
"【": "",
"】": "",
"[": "",
"]": "",
"—": "",
"～": "-",
"~": "-",
"「": "",
"」": "",
"¿": "",
"¡": "",
}
def collapse_whitespace(text):
# Regular expression matching whitespace:
_whitespace_re = re.compile(r"\s+")
return re.sub(_whitespace_re, " ", text).strip()
def remove_punctuation_at_begin(text):
return re.sub(r"^[,.!?]+", "", text)
def remove_aux_symbols(text):
text = re.sub(r"[\<\>\(\)\[\]\"\«\»]+", "", text)
return text
def replace_symbols(text):
text = text.replace(";", ",")
text = text.replace("-", " ")
text = text.replace(":", ",")
text = text.replace("&", " et ")
return text
def expand_abbreviations(text):
for regex, replacement in _abbreviations:
text = re.sub(regex, replacement, text)
return text
def replace_punctuation(text):
pattern = re.compile("|".join(re.escape(p) for p in rep_map.keys()))
replaced_text = pattern.sub(lambda x: rep_map[x.group()], text)
return replaced_text
def text_normalize(text):
text = expand_abbreviations(text)
text = replace_punctuation(text)
text = replace_symbols(text)
text = remove_aux_symbols(text)
text = remove_punctuation_at_begin(text)
text = collapse_whitespace(text)
text = re.sub(r"([^\.,!\?\-…])$", r"\1", text)
return text
def french_to_ipa(text, text_tokenizer):
if type(text) == str:
text = text_normalize(text)
phonemes = text_tokenizer(text)
return phonemes
else:
for i, t in enumerate(text):
text[i] = text_normalize(t)
return text_tokenizer(text)
|