File size: 6,964 Bytes
ca90f09
d02ad9c
 
 
194fffd
 
0d1adfa
 
194fffd
f74bce2
ee48acc
f74bce2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
d6aee2b
 
 
 
 
 
 
 
e886026
939c1fe
 
 
 
 
d636635
ee48acc
 
f74bce2
 
 
 
194fffd
 
fc50d18
194fffd
e1c65f1
e7a921a
533ef97
e886026
533ef97
e886026
533ef97
e886026
533ef97
e1c65f1
 
 
 
 
 
 
3444a7f
 
 
 
 
0dfedcd
d636635
f74bce2
d636635
 
f74bce2
 
939c1fe
d636635
0dfedcd
f74bce2
0dfedcd
 
f74bce2
 
939c1fe
0dfedcd
e886026
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
d636635
f74bce2
d636635
 
f74bce2
 
 
 
 
 
 
 
 
 
939c1fe
d636635
e886026
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0dfedcd
 
e886026
 
194fffd
 
 
142fdc7
 
 
19d535a
142fdc7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
e886026
142fdc7
c6f8b29
142fdc7
 
0dfedcd
142fdc7
 
 
262bc3f
142fdc7
939c1fe
 
f74bce2
 
 
 
 
939c1fe
 
58c1f3f
939c1fe
142fdc7
194fffd
 
98d97a5
2cd6c69
194fffd
 
cff1c92
3444a7f
0f6c489
e886026
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
import sys
import os

import gradio as gr
from TTS.api import TTS

tts = TTS("tts_models/multilingual/multi-dataset/xtts_v1")
tts.to("cuda")

def predict(prompt, language, audio_file_pth, mic_file_path, use_mic, agree):
    if agree == True:
        if use_mic == True:
            if mic_file_path is not None:
                speaker_wav=mic_file_path
            else:
                gr.Warning("Please record your voice with Microphone, or uncheck Use Microphone to use reference audios")
                return (
                    None,
                    None,
                ) 
                
        else:
            speaker_wav=audio_file_pth

        if len(prompt)<2:
            gr.Warning("Please give a longer prompt text")
            return (
                    None,
                    None,
                )
        try:   
            tts.tts_to_file(
                text=prompt,
                file_path="output.wav",
                speaker_wav=speaker_wav,
                language=language,
            )
        except RuntimeError as e :
            if "device-assert" in str(e):
                # cannot do anything on cuda device side error, need tor estart
                gr.Warning("Unhandled Exception encounter, please retry in a minute")
                print("Cuda device-assert Runtime encountered need restart")
                sys.exit("Exit due to cuda device-assert")
            else:
                raise e
            
        return (
            gr.make_waveform(
                audio="output.wav",
            ),
            "output.wav",
        )
    else:
        gr.Warning("Please accept the Terms & Condition!")
        return (
                None,
                None,
            ) 


title = "Coqui🐸 XTTS"

description = """
<a href="https://huggingface.co/coqui/XTTS-v1">XTTS</a> is a Voice generation model that lets you clone voices into different languages by using just a quick 3-second audio clip. 
<br/>
XTTS is built on previous research, like Tortoise, with additional architectural innovations and training to make cross-language voice cloning and multilingual speech generation possible. 
<br/>
This is the same model that powers our creator application <a href="https://coqui.ai">Coqui Studio</a> as well as the <a href="https://docs.coqui.ai">Coqui API</a>. In production we apply modifications to make low-latency streaming possible.
<br/>
Leave a star on the Github <a href="https://github.com/coqui-ai/TTS">🐸TTS</a>, where our open-source inference and training code lives.
<br/>
<p>For faster inference without waiting in the queue, you should duplicate this space and upgrade to GPU via the settings.
<br/>
<a href="https://huggingface.co/spaces/coqui/xtts?duplicate=true">
<img style="margin-top: 0em; margin-bottom: 0em" src="https://bit.ly/3gLdBN6" alt="Duplicate Space"></a>
</p>
"""

article = """
<div style='margin:20px auto;'>
<p>By using this demo you agree to the terms of the Coqui Public Model License at https://coqui.ai/cpml</p>
</div>
"""
examples = [
    [
        "Once when I was six years old I saw a magnificent picture",
        "en",
        "examples/female.wav",
        None,
        False,
        True,
    ],
    [
        "Lorsque j'avais six ans j'ai vu, une fois, une magnifique image",
        "fr",
        "examples/male.wav",
        None,
        False,
        True,
    ],
    [
        "Als ich sechs war, sah ich einmal ein wunderbares Bild",
        "de",
        "examples/female.wav",
        None,
        False,
        True,
    ],
    [
        "Cuando tenía seis años, vi una vez una imagen magnífica",
        "es",
        "examples/male.wav",
        None,
        False,
        True,
    ],
    [
        "Quando eu tinha seis anos eu vi, uma vez, uma imagem magnífica",
        "pt",
        "examples/female.wav",
        None,
        False,
        True,
    ],
    [
        "Kiedy miałem sześć lat, zobaczyłem pewnego razu wspaniały obrazek",
        "pl",
        "examples/male.wav",
        None,
        False,
        True,
    ],
    [
        "Un tempo lontano, quando avevo sei anni, vidi un magnifico disegno",
        "it",
        "examples/female.wav",
        None,
        False,
        True,
    ],
    [
        "Bir zamanlar, altı yaşındayken, muhteşem bir resim gördüm",
        "tr",
        "examples/female.wav",
        None,
        False,
        True,
    ],
    [
        "Когда мне было шесть лет, я увидел однажды удивительную картинку",
        "ru",
        "examples/female.wav",
        None,
        False,
        True,
    ],
    [
        "Toen ik een jaar of zes was, zag ik op een keer een prachtige plaat",
        "nl",
        "examples/male.wav",
        None,
        False,
        True,
    ],
    [
        "Když mi bylo šest let, viděl jsem jednou nádherný obrázek",
        "cs",
        "examples/female.wav",
        None,
        False,
        True,
    ],
    [
        "当我还只有六岁的时候, 看到了一副精彩的插画",
        "zh-cn",
        "examples/female.wav",
        None,
        False,
        True,
    ],
]



gr.Interface(
    fn=predict,
    inputs=[
        gr.Textbox(
            label="Text Prompt",
            info="One or two sentences at a time is better",
            value="Hi there, I'm your new voice clone. Try your best to upload quality audio. Like my grandma used to say: 'Garbage in, garbage out!'",
        ),
        gr.Dropdown(
            label="Language",
            info="Select an output language for the synthesised speech",
            choices=[
                "en",
                "es",
                "fr",
                "de",
                "it",
                "pt",
                "pl",
                "tr",
                "ru",
                "nl",
                "cs",
                "ar",
                "zh-cn",
            ],
            max_choices=1,
            value="en",
        ),
        gr.Audio(
            label="Reference Audio",
            info="Click on the ✎ button to upload your own target speaker audio",
            type="filepath",
            value="examples/female.wav",
        ),
        gr.Audio(source="microphone",
                 type="filepath",
                 info="Use your microphone to record audio",
                 label="Use Microphone for Reference"),
        gr.Checkbox(label="Check to use Microphone as Reference", value=False),
        gr.Checkbox(
            label="Agree",
            value=False,
            info="I agree to the terms of the Coqui Public Model License at https://coqui.ai/cpml",
        ),
    ],
    outputs=[
        gr.Video(label="Waveform Visual"),
        gr.Audio(label="Synthesised Audio"),
    ],
    title=title,
    description=description,
    article=article,
    examples=examples,
).queue().launch(debug=True)