File size: 7,781 Bytes
890de26
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
# -*- coding:utf-8 -*-
# @FileName  :postprocee.py
# @Time      :2023/8/8 20:26
# @Author    :lovemefan
# @Email     :[email protected]
# Copyright FunASR (https://github.com/alibaba-damo-academy/FunASR). All Rights Reserved.
#  MIT License  (https://opensource.org/licenses/MIT)

from typing import Any, List, Union


def isChinese(ch: str):
    if "\u4e00" <= ch <= "\u9fff" or "\u0030" <= ch <= "\u0039":
        return True
    return False


def isAllChinese(word: Union[List[Any], str]):
    word_lists = []
    for i in word:
        cur = i.replace(" ", "")
        cur = cur.replace("</s>", "")
        cur = cur.replace("<s>", "")
        word_lists.append(cur)

    if len(word_lists) == 0:
        return False

    for ch in word_lists:
        if isChinese(ch) is False:
            return False
    return True


def isAllAlpha(word: Union[List[Any], str]):
    word_lists = []
    for i in word:
        cur = i.replace(" ", "")
        cur = cur.replace("</s>", "")
        cur = cur.replace("<s>", "")
        word_lists.append(cur)

    if len(word_lists) == 0:
        return False

    for ch in word_lists:
        if ch.isalpha() is False and ch != "'":
            return False
        elif ch.isalpha() is True and isChinese(ch) is True:
            return False

    return True


# def abbr_dispose(words: List[Any]) -> List[Any]:
def abbr_dispose(words: List[Any], time_stamp: List[List] = None) -> List[Any]:
    words_size = len(words)
    word_lists = []
    abbr_begin = []
    abbr_end = []
    last_num = -1
    ts_lists = []
    ts_nums = []
    ts_index = 0
    for num in range(words_size):
        if num <= last_num:
            continue

        if len(words[num]) == 1 and words[num].encode("utf-8").isalpha():
            if (
                num + 1 < words_size
                and words[num + 1] == " "
                and num + 2 < words_size
                and len(words[num + 2]) == 1
                and words[num + 2].encode("utf-8").isalpha()
            ):
                # found the begin of abbr
                abbr_begin.append(num)
                num += 2
                abbr_end.append(num)
                # to find the end of abbr
                while True:
                    num += 1
                    if num < words_size and words[num] == " ":
                        num += 1
                        if (
                            num < words_size
                            and len(words[num]) == 1
                            and words[num].encode("utf-8").isalpha()
                        ):
                            abbr_end.pop()
                            abbr_end.append(num)
                            last_num = num
                        else:
                            break
                    else:
                        break

    for num in range(words_size):
        if words[num] == " ":
            ts_nums.append(ts_index)
        else:
            ts_nums.append(ts_index)
            ts_index += 1
    last_num = -1
    for num in range(words_size):
        if num <= last_num:
            continue

        if num in abbr_begin:
            if time_stamp is not None:
                begin = time_stamp[ts_nums[num]][0]
            word_lists.append(words[num].upper())
            num += 1
            while num < words_size:
                if num in abbr_end:
                    word_lists.append(words[num].upper())
                    last_num = num
                    break
                else:
                    if words[num].encode("utf-8").isalpha():
                        word_lists.append(words[num].upper())
                num += 1
            if time_stamp is not None:
                end = time_stamp[ts_nums[num]][1]
                ts_lists.append([begin, end])
        else:
            word_lists.append(words[num])
            if time_stamp is not None and words[num] != " ":
                begin = time_stamp[ts_nums[num]][0]
                end = time_stamp[ts_nums[num]][1]
                ts_lists.append([begin, end])
                begin = end

    if time_stamp is not None:
        return word_lists, ts_lists
    else:
        return word_lists


def sentence_postprocess(words: List[Any], time_stamp: List[List] = None):
    middle_lists = []
    word_lists = []
    word_item = ""
    ts_lists = []

    # wash words lists
    for i in words:
        word = ""
        if isinstance(i, str):
            word = i
        else:
            word = i.decode("utf-8")

        if word in ["<s>", "</s>", "<unk>"]:
            continue
        else:
            middle_lists.append(word)

    # all chinese characters
    if isAllChinese(middle_lists):
        for i, ch in enumerate(middle_lists):
            word_lists.append(ch.replace(" ", ""))
        if time_stamp is not None:
            ts_lists = time_stamp

    # all alpha characters
    elif isAllAlpha(middle_lists):
        ts_flag = True
        for i, ch in enumerate(middle_lists):
            if ts_flag and time_stamp is not None:
                begin = time_stamp[i][0]
                end = time_stamp[i][1]
            word = ""
            if "@@" in ch:
                word = ch.replace("@@", "")
                word_item += word
                if time_stamp is not None:
                    ts_flag = False
                    end = time_stamp[i][1]
            else:
                word_item += ch
                word_lists.append(word_item)
                word_lists.append(" ")
                word_item = ""
                if time_stamp is not None:
                    ts_flag = True
                    end = time_stamp[i][1]
                    ts_lists.append([begin, end])
                    begin = end

    # mix characters
    else:
        alpha_blank = False
        ts_flag = True
        begin = -1
        end = -1
        for i, ch in enumerate(middle_lists):
            if ts_flag and time_stamp is not None:
                begin = time_stamp[i][0]
                end = time_stamp[i][1]
            word = ""
            if isAllChinese(ch):
                if alpha_blank is True:
                    word_lists.pop()
                word_lists.append(ch)
                alpha_blank = False
                if time_stamp is not None:
                    ts_flag = True
                    ts_lists.append([begin, end])
                    begin = end
            elif "@@" in ch:
                word = ch.replace("@@", "")
                word_item += word
                alpha_blank = False
                if time_stamp is not None:
                    ts_flag = False
                    end = time_stamp[i][1]
            elif isAllAlpha(ch):
                word_item += ch
                word_lists.append(word_item)
                word_lists.append(" ")
                word_item = ""
                alpha_blank = True
                if time_stamp is not None:
                    ts_flag = True
                    end = time_stamp[i][1]
                    ts_lists.append([begin, end])
                    begin = end
            else:
                raise ValueError("invalid character: {}".format(ch))

    if time_stamp is not None:
        word_lists, ts_lists = abbr_dispose(word_lists, ts_lists)
        real_word_lists = []
        for ch in word_lists:
            if ch != " ":
                real_word_lists.append(ch)
        sentence = " ".join(real_word_lists).strip()
        return sentence, ts_lists, real_word_lists
    else:
        word_lists = abbr_dispose(word_lists)
        real_word_lists = []
        for ch in word_lists:
            if ch != " ":
                real_word_lists.append(ch)
        sentence = "".join(word_lists).strip()
        return sentence, real_word_lists