File size: 4,903 Bytes
aeee641
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
#!/usr/bin/env python3
# Copyright    2023  Xiaomi Corp.        (authors: Fangjun Kuang)

from typing import Dict, List

import onnxruntime
import soundfile
import torch


def display(sess):
    for i in sess.get_inputs():
        print(i)

    print("-" * 10)
    for o in sess.get_outputs():
        print(o)


class OnnxModel:
    def __init__(
        self,
        model: str,
    ):
        session_opts = onnxruntime.SessionOptions()
        session_opts.inter_op_num_threads = 1
        session_opts.intra_op_num_threads = 4

        self.session_opts = session_opts

        self.model = onnxruntime.InferenceSession(
            model,
            sess_options=self.session_opts,
        )
        display(self.model)

        meta = self.model.get_modelmeta().custom_metadata_map
        self.add_blank = int(meta["add_blank"])
        self.sample_rate = int(meta["sample_rate"])
        self.punctuation = meta["punctuation"].split()
        print(meta)

    def __call__(
        self,
        x: torch.Tensor,
        sid: int,
    ) -> torch.Tensor:
        """
        Args:
          x:
            A int64 tensor of shape (L,)
        """
        x = x.unsqueeze(0)
        x_length = torch.tensor([x.shape[1]], dtype=torch.int64)
        noise_scale = torch.tensor([1], dtype=torch.float32)
        length_scale = torch.tensor([1], dtype=torch.float32)
        noise_scale_w = torch.tensor([1], dtype=torch.float32)
        sid = torch.tensor([sid], dtype=torch.int64)

        y = self.model.run(
            [
                self.model.get_outputs()[0].name,
            ],
            {
                self.model.get_inputs()[0].name: x.numpy(),
                self.model.get_inputs()[1].name: x_length.numpy(),
                self.model.get_inputs()[2].name: noise_scale.numpy(),
                self.model.get_inputs()[3].name: length_scale.numpy(),
                self.model.get_inputs()[4].name: noise_scale_w.numpy(),
                self.model.get_inputs()[5].name: sid.numpy(),
            },
        )[0]
        return torch.from_numpy(y).squeeze()


def read_lexicon() -> Dict[str, List[str]]:
    ans = dict()
    with open("./lexicon.txt", encoding="utf-8") as f:
        for line in f:
            w_p = line.split()
            w = w_p[0]
            p = w_p[1:]
            ans[w] = p
    return ans


def read_tokens() -> Dict[str, int]:
    ans = dict()
    with open("./tokens.txt", encoding="utf-8") as f:
        for line in f:
            t_i = line.strip().split()
            if len(t_i) == 1:
                token = " "
                idx = t_i[0]
            else:
                assert len(t_i) == 2, (t_i, line)
                token = t_i[0]
                idx = t_i[1]
            ans[token] = int(idx)
    return ans


def convert_lexicon(lexicon, tokens):
    for w in lexicon:
        phones = lexicon[w]
        try:
            p = [tokens[i] for i in phones]
            lexicon[w] = p
        except Exception:
            #  print("skip", w)
            continue


"""
skip rapprochement
skip croissants
skip aix-en-provence
skip provence
skip croissant
skip denouement
skip hola
skip blanc
"""


def get_text(text, lexicon, tokens, punctuation):
    text = text.lower().split()
    ans = []
    for i in range(len(text)):
        w = text[i]
        punct = None

        if w[0] in punctuation:
            ans.append(tokens[w[0]])
            w = w[1:]

        if w[-1] in punctuation:
            punct = tokens[w[-1]]
            w = w[:-1]

        if w in lexicon:
            ans.extend(lexicon[w])
            if punct:
                ans.append(punct)

            if i != len(text) - 1:
                ans.append(tokens[" "])
            continue
        print("ignore", w)
    return ans


def generate(model, text, lexicon, tokens, sid):
    x = get_text(
        text,
        lexicon,
        tokens,
        model.punctuation,
    )
    if model.add_blank:
        x2 = [0] * (2 * len(x) + 1)
        x2[1::2] = x
        x = x2

    x = torch.tensor(x, dtype=torch.int64)

    y = model(x, sid=sid)

    return y


def main():
    model = OnnxModel("./vits-vctk.onnx")

    lexicon = read_lexicon()
    tokens = read_tokens()
    convert_lexicon(lexicon, tokens)

    text = "Liliana, our most beautiful and lovely assistant"
    y = generate(model, text, lexicon, tokens, sid=0)
    soundfile.write("test-0.wav", y.numpy(), model.sample_rate)

    text = "Ask not what your country can do for you; ask what you can do for your country."
    y = generate(model, text, lexicon, tokens, sid=1)
    soundfile.write("test-1.wav", y.numpy(), model.sample_rate)

    text = "Success is not final, failure is not fatal, it is the courage to continue that counts!"
    y = generate(model, text, lexicon, tokens, sid=2)
    soundfile.write("test-2.wav", y.numpy(), model.sample_rate)


if __name__ == "__main__":
    main()