File size: 4,197 Bytes
1cf1e13
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
import pickle
import os
import re
from g2p_en import G2p

from text import symbols

current_file_path = os.path.dirname(__file__)
CMU_DICT_PATH = os.path.join(current_file_path, "cmudict.rep")
CACHE_PATH = os.path.join(current_file_path, "cmudict_cache.pickle")
_g2p = G2p()

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",
}


def post_replace_ph(ph):
    rep_map = {
        ":": ",",
        ";": ",",
        ",": ",",
        "。": ".",
        "!": "!",
        "?": "?",
        "\n": ".",
        "·": ",",
        "、": ",",
        "...": "…",
        "v": "V",
    }
    if ph in rep_map.keys():
        ph = rep_map[ph]
    if ph in symbols:
        return ph
    if ph not in symbols:
        ph = "UNK"
    return ph


def read_dict():
    g2p_dict = {}
    start_line = 49
    with open(CMU_DICT_PATH) as f:
        line = f.readline()
        line_index = 1
        while line:
            if line_index >= start_line:
                line = line.strip()
                word_split = line.split("  ")
                word = word_split[0]

                syllable_split = word_split[1].split(" - ")
                g2p_dict[word] = []
                for syllable in syllable_split:
                    phone_split = syllable.split(" ")
                    g2p_dict[word].append(phone_split)

            line_index = line_index + 1
            line = f.readline()

    return g2p_dict


def cache_dict(g2p_dict, file_path):
    with open(file_path, "wb") as pickle_file:
        pickle.dump(g2p_dict, pickle_file)


def get_dict():
    if os.path.exists(CACHE_PATH):
        with open(CACHE_PATH, "rb") as pickle_file:
            g2p_dict = pickle.load(pickle_file)
    else:
        g2p_dict = read_dict()
        cache_dict(g2p_dict, CACHE_PATH)

    return g2p_dict


eng_dict = get_dict()


def refine_ph(phn):
    tone = 0
    if re.search(r"\d$", phn):
        tone = int(phn[-1]) + 1
        phn = phn[:-1]
    return phn.lower(), tone


def refine_syllables(syllables):
    tones = []
    phonemes = []
    for phn_list in syllables:
        for i in range(len(phn_list)):
            phn = phn_list[i]
            phn, tone = refine_ph(phn)
            phonemes.append(phn)
            tones.append(tone)
    return phonemes, tones


def text_normalize(text):
    # todo: eng text normalize
    return text


def g2p(text):
    phones = []
    tones = []
    words = re.split(r"([,;.\-\?\!\s+])", text)
    for w in words:
        if w.upper() in eng_dict:
            phns, tns = refine_syllables(eng_dict[w.upper()])
            phones += phns
            tones += tns
        else:
            phone_list = list(filter(lambda p: p != " ", _g2p(w)))
            for ph in phone_list:
                if ph in arpa:
                    ph, tn = refine_ph(ph)
                    phones.append(ph)
                    tones.append(tn)
                else:
                    phones.append(ph)
                    tones.append(0)
    # todo: implement word2ph
    word2ph = [1 for i in phones]

    phones = [post_replace_ph(i) for i in phones]
    return phones, tones, word2ph


if __name__ == "__main__":
    # print(get_dict())
    # print(eng_word_to_phoneme("hello"))
    print(g2p("In this paper, we propose 1 DSPGAN, a GAN-based universal vocoder."))
    # all_phones = set()
    # for k, syllables in eng_dict.items():
    #     for group in syllables:
    #         for ph in group:
    #             all_phones.add(ph)
    # print(all_phones)