Update app.py
Browse files
app.py
CHANGED
@@ -1,483 +1,572 @@
|
|
1 |
import gradio as gr
|
2 |
-
import
|
3 |
-
import tempfile
|
4 |
-
import asyncio
|
5 |
-
import traceback
|
6 |
-
import os
|
7 |
-
import google.generativeai as genai
|
8 |
-
import logging
|
9 |
-
import time
|
10 |
-
import threading
|
11 |
-
import sys
|
12 |
-
import base64
|
13 |
import mimetypes
|
|
|
14 |
import re
|
15 |
import struct
|
16 |
-
|
17 |
-
|
18 |
-
from
|
19 |
-
|
20 |
-
# --- پیکربندی لاگینگ ---
|
21 |
-
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(threadName)s - %(message)s')
|
22 |
-
|
23 |
-
# --- چرخش کلیدهای Gemini API ---
|
24 |
-
API_KEYS_GEMINI = []
|
25 |
-
i = 1
|
26 |
-
while True:
|
27 |
-
key = os.environ.get(f'GEMINI_API_KEY_{i}')
|
28 |
-
if key:
|
29 |
-
API_KEYS_GEMINI.append(key)
|
30 |
-
i += 1
|
31 |
-
else:
|
32 |
-
break
|
33 |
-
|
34 |
-
NUM_GEMINI_KEYS = len(API_KEYS_GEMINI)
|
35 |
-
current_gemini_key_index = 0
|
36 |
-
gemini_operations_lock = asyncio.Lock()
|
37 |
-
|
38 |
-
if NUM_GEMINI_KEYS == 0:
|
39 |
-
logging.error(
|
40 |
-
'خطای حیاتی: هیچ Secret با نام GEMINI_API_KEY_n (مثلاً GEMINI_API_KEY_1) یافت نشد! '
|
41 |
-
'قابلیت ترجمه غیرفعال خواهد بود. لطفاً Secret ها را در تنظیمات Space خود اضافه کنید.'
|
42 |
-
)
|
43 |
-
else:
|
44 |
-
logging.info(f"تعداد {NUM_GEMINI_KEYS} کلید API جیمینای بارگذاری شد.")
|
45 |
-
|
46 |
-
# --- ریستارت خودکار ---
|
47 |
-
def auto_restart_service():
|
48 |
-
RESTART_INTERVAL_SECONDS = 24 * 60 * 60 # 24 ساعت
|
49 |
-
logging.info(f"سرویس برای ریاستارت خودکار پس از {RESTART_INTERVAL_SECONDS / 3600:.0f} ساعت زمانبندی شده است.")
|
50 |
-
time.sleep(RESTART_INTERVAL_SECONDS)
|
51 |
-
logging.info(f"زمان ریاستارت خودکار ({RESTART_INTERVAL_SECONDS / 3600:.0f} ساعت) فرا رسیده است. برنامه خارج میشود تا توسط پلتفرم ریاستارت شود...")
|
52 |
-
os._exit(1)
|
53 |
-
|
54 |
-
# --- دیکشنری صداهای انگلیسی ---
|
55 |
-
language_dict_persian_keys = {
|
56 |
-
'انگلیسی (آمریکا) - جنی (زن)': 'en-US-JennyNeural',
|
57 |
-
'انگلیسی (آمریکا) - گای (مرد)': 'en-US-GuyNeural',
|
58 |
-
'انگلیسی (آمریکا) - آنا (زن، صدای کودک)': 'en-US-AnaNeural',
|
59 |
-
'انگلیسی (آمریکا) - آریا (زن)': 'en-US-AriaNeural',
|
60 |
-
'انگلیسی (آمریکا) - کریستوفر (مرد)': 'en-US-ChristopherNeural',
|
61 |
-
'انگلیسی (آمریکا) - اریک (مرد)': 'en-US-EricNeural',
|
62 |
-
'انگلیسی (آمریکا) - میشل (زن)': 'en-US-MichelleNeural',
|
63 |
-
'انگلیسی (آمریکا) - راجر (مرد)': 'en-US-RogerNeural',
|
64 |
-
'انگلیسی (بریتانیا) - لیبی (زن)': 'en-GB-LibbyNeural',
|
65 |
-
'انگلیسی (بریتانیا) - میزی (زن)': 'en-GB-MaisieNeural',
|
66 |
-
'انگلیسی (بریتانیا) - رایان (مرد)': 'en-GB-RyanNeural',
|
67 |
-
'انگلیسی (بریتانیا) - سونیا (زن)': 'en-GB-SoniaNeural',
|
68 |
-
'انگلیسی (بریتانیا) - توماس (مرد)': 'en-GB-ThomasNeural',
|
69 |
-
'انگلیسی (بریتانیا) - میا (زن، جدید)': 'en-GB-MiaNeural',
|
70 |
-
'انگلیسی (استرالیا) - ناتاشا (زن)': 'en-AU-NatashaNeural',
|
71 |
-
'انگلیسی (استرالیا) - ویلیام (مرد)': 'en-AU-WilliamNeural',
|
72 |
-
'انگلیسی (کانادا) - کلارا (زن)': 'en-CA-ClaraNeural',
|
73 |
-
'انگلیسی (کانادا) - لیام (مرد)': 'en-CA-LiamNeural',
|
74 |
-
'انگلیسی (ایرلند) - امیلی (زن)': 'en-IE-EmilyNeural',
|
75 |
-
'انگلیسی (ایرلند) - کانر (مرد)': 'en-IE-ConnorNeural',
|
76 |
-
'انگلیسی (هند) - نیرجا (زن)': 'en-IN-NeerjaNeural',
|
77 |
-
'انگلیسی (هند) - پرابهات (مرد)': 'en-IN-PrabhatNeural',
|
78 |
-
'انگلیسی (آفریقای جنوبی) - لیا (زن)': 'en-ZA-LeahNeural',
|
79 |
-
'انگلیسی (آفریقای جنوبی) - لوک (مرد)': 'en-ZA-LukeNeural',
|
80 |
-
'انگلیسی (هنگ کنگ) - یان (زن)': 'en-HK-YanNeural',
|
81 |
-
'انگلیسی (هنگ کنگ) - سم (مرد)': 'en-HK-SamNeural',
|
82 |
-
'انگلیسی (نیوزیلند) - میچل (مرد)': 'en-NZ-MitchellNeural',
|
83 |
-
'انگلیسی (فیلیپین) - روزا (زن)': 'en-PH-RosaNeural',
|
84 |
-
'انگلیسی (فیلیپین) - جیمز (مرد)': 'en-PH-JamesNeural',
|
85 |
-
'انگلیسی (سنگاپور) - لونا (زن)': 'en-SG-LunaNeural',
|
86 |
-
'انگلیسی (سنگاپور) - وین (مرد)': 'en-SG-WayneNeural',
|
87 |
-
'انگلیسی (کنیا) - آسیلیا (زن)': 'en-KE-AsiliaNeural',
|
88 |
-
'انگلیسی (کنیا) - چیلمبا (مرد)': 'en-KE-ChilembaNeural',
|
89 |
-
'انگلیسی (نیجریه) - ازینه (زن)': 'en-NG-EzinneNeural',
|
90 |
-
'انگلیسی (نیجریه) - آبئو (مرد)': 'en-NG-AbeoNeural',
|
91 |
-
'انگلیسی (تانزانیا) - ایمانی (زن)': 'en-TZ-ImaniNeural',
|
92 |
-
'انگلیسی (تانزانیا) - الیمو (مرد)': 'en-TZ-ElimuNeural',
|
93 |
-
}
|
94 |
-
|
95 |
-
# --- توابع اصلی ---
|
96 |
-
async def translate_text_gemini(text, target_language="English"):
|
97 |
-
if NUM_GEMINI_KEYS == 0:
|
98 |
-
return "خطا: سرویس ترجمه پیکربندی نشده (هیچ کلید API جیمینای معتبری یافت نشد).", None
|
99 |
-
if not text or not text.strip():
|
100 |
-
return "خطا: متنی برای ترجمه وارد نشده است.", None
|
101 |
-
|
102 |
-
selected_api_key = None
|
103 |
-
model_to_use = 'gemini-1.5-flash-latest'
|
104 |
-
|
105 |
-
async with gemini_operations_lock:
|
106 |
-
global current_gemini_key_index
|
107 |
-
key_index_to_use = current_gemini_key_index
|
108 |
-
selected_api_key = API_KEYS_GEMINI[key_index_to_use]
|
109 |
-
current_gemini_key_index = (current_gemini_key_index + 1) % NUM_GEMINI_KEYS
|
110 |
-
logging.info(f"قفل Gemini گرفته شد. استفاده از کلید API با اندیس: {key_index_to_use} (...{selected_api_key[-4:]}) برای مدل {model_to_use}")
|
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 |
try:
|
144 |
-
|
145 |
-
|
146 |
-
|
147 |
-
|
148 |
-
|
149 |
-
|
150 |
-
|
151 |
-
|
152 |
-
|
153 |
-
|
154 |
-
|
155 |
-
|
156 |
-
|
157 |
-
|
158 |
-
|
159 |
-
|
160 |
-
|
161 |
-
|
|
|
|
|
|
|
|
|
|
|
162 |
|
163 |
-
|
|
|
164 |
|
165 |
-
|
166 |
-
if
|
167 |
-
|
168 |
-
|
169 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
170 |
|
171 |
-
|
172 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
173 |
|
174 |
-
|
175 |
-
|
|
|
|
|
176 |
|
177 |
-
if english_tts_voice_key not in language_dict_persian_keys:
|
178 |
-
if language_dict_persian_keys:
|
179 |
-
english_tts_voice_key = list(language_dict_persian_keys.keys())[0]
|
180 |
-
logging.warning(f"صدای انتخابی نامعتبر، به پیشفرض '{english_tts_voice_key}' تغییر یافت.")
|
181 |
-
else:
|
182 |
-
return f"{translated_text_output}\n\n(خطای TTS: هیچ صدایی موجود نیست.)", None
|
183 |
|
184 |
-
|
185 |
-
|
186 |
-
|
187 |
-
|
188 |
-
|
189 |
-
|
190 |
-
|
191 |
-
#
|
192 |
-
|
193 |
-
|
194 |
-
|
195 |
-
|
196 |
-
|
197 |
-
|
198 |
-
|
199 |
-
|
200 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
201 |
@import url('https://fonts.googleapis.com/css2?family=Poppins:wght@400;500;600;700;800&display=swap');
|
|
|
|
|
202 |
:root {{
|
203 |
-
--
|
204 |
-
--
|
205 |
-
--
|
206 |
-
--
|
207 |
-
--
|
208 |
-
--
|
209 |
-
--
|
210 |
-
--
|
211 |
-
--
|
212 |
-
--
|
213 |
-
--
|
214 |
-
|
215 |
-
--radius-
|
216 |
-
--shadow-
|
217 |
-
--shadow-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
218 |
}}
|
219 |
-
|
220 |
-
|
|
|
221 |
direction: rtl;
|
222 |
-
background: var(--
|
223 |
-
color: var(--
|
224 |
-
|
225 |
-
|
226 |
}}
|
227 |
-
|
228 |
-
|
229 |
-
min-height:100vh;
|
230 |
-
margin:0 !important;
|
231 |
-
|
232 |
-
display:flex;
|
233 |
-
flex-direction:column;
|
234 |
}}
|
235 |
-
|
236 |
-
|
237 |
-
|
238 |
-
|
239 |
-
|
240 |
-
border-bottom-left-radius:
|
241 |
-
border-bottom-right-radius:
|
242 |
-
|
243 |
}}
|
244 |
-
.app-header-
|
245 |
-
|
246 |
-
|
247 |
-
|
248 |
-
text-shadow: 0 2px 4px rgba(0,0,0,0.15);
|
249 |
}}
|
250 |
-
.app-header-
|
251 |
-
font-size:
|
252 |
-
|
253 |
-
margin-top:0;
|
254 |
-
opacity: 0.9;
|
255 |
}}
|
256 |
-
.
|
257 |
-
|
258 |
-
|
259 |
-
|
260 |
-
|
261 |
-
|
262 |
-
|
263 |
-
|
264 |
-
position:relative;
|
265 |
-
z-index:10;
|
266 |
}}
|
267 |
-
|
268 |
-
|
269 |
-
|
270 |
-
|
271 |
-
|
272 |
-
|
273 |
-
|
274 |
-
|
275 |
-
|
276 |
-
|
277 |
-
|
278 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
279 |
}}
|
280 |
-
|
281 |
-
|
282 |
-
|
283 |
-
|
284 |
-
|
285 |
-
|
286 |
-
|
287 |
-
|
288 |
-
|
289 |
-
|
290 |
-
|
291 |
-
|
292 |
-
width:100%;
|
293 |
-
margin-top:1.5rem !important;
|
294 |
}}
|
295 |
-
.
|
296 |
-
|
297 |
-
transform:
|
298 |
-
box-shadow:
|
299 |
}}
|
300 |
-
|
301 |
-
|
302 |
-
.gr-
|
303 |
-
|
304 |
-
|
305 |
-
|
306 |
-
background-color: var(--app-input-bg) !important;
|
307 |
-
box-shadow: inset 0 1px 2px rgba(0,0,0,0.05);
|
308 |
-
padding: 0.75rem !important;
|
309 |
}}
|
310 |
-
|
311 |
-
.
|
312 |
-
|
313 |
-
border-color: var(--app-button-bg) !important;
|
314 |
-
box-shadow: 0 0 0 3px rgba(41,121,255,0.2) !important;
|
315 |
}}
|
316 |
-
|
317 |
-
|
318 |
-
|
319 |
-
|
320 |
-
margin-
|
321 |
}}
|
322 |
-
|
323 |
-
|
324 |
-
|
325 |
-
|
|
|
|
|
|
|
326 |
}}
|
327 |
-
.
|
328 |
-
|
329 |
-
|
330 |
-
|
331 |
-
|
332 |
-
|
333 |
-
|
334 |
-
text-align:right;
|
335 |
}}
|
336 |
-
|
337 |
-
|
338 |
-
|
339 |
-
label[for="volume_slider"] > .label-text::before {{ content: '🔊'; margin-left: 8px; }}
|
340 |
-
label[for="pitch_slider"] > .label-text::before {{ content: '🎵'; margin-left: 8px; }}
|
341 |
-
label[for="translated_text_output"] > .label-text::before {{ content: '📜'; margin-left: 8px; }}
|
342 |
-
label[for="output_audio"] > .label-text::before {{ content: '🎧'; margin-left: 8px; }}
|
343 |
-
|
344 |
-
#output_audio_player audio {{
|
345 |
-
width: 100%;
|
346 |
-
border-radius: var(--radius-input);
|
347 |
-
margin-top:0.8rem;
|
348 |
}}
|
349 |
-
|
350 |
-
|
351 |
-
|
352 |
-
|
353 |
-
|
354 |
-
|
355 |
-
|
356 |
-
|
357 |
}}
|
358 |
-
|
359 |
-
|
360 |
-
|
361 |
-
|
362 |
-
|
|
|
|
|
|
|
|
|
363 |
}}
|
364 |
"""
|
365 |
|
366 |
-
# ---
|
367 |
-
|
368 |
-
|
369 |
-
|
370 |
-
|
371 |
-
|
372 |
-
|
373 |
-
|
374 |
-
|
375 |
-
|
376 |
-
|
377 |
-
<div class='app-header-alpha'>
|
378 |
-
<h1>Alpha Translator</h1>
|
379 |
-
<p>جادوی ترجمه و تلفظ در دستان شما</p>
|
380 |
-
</div>
|
381 |
""")
|
382 |
|
383 |
-
|
384 |
-
|
385 |
-
|
386 |
-
|
387 |
-
|
388 |
-
|
389 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
390 |
)
|
391 |
-
gr.Markdown(f"<div style='background-color:#FFFBEB; color:#92400E; padding:10px 12px; border-radius:8px; border:1px solid #FDE68A; text-align:center; margin-bottom:1rem;'>{missing_key_msg}</div>")
|
392 |
-
|
393 |
-
with gr.Row():
|
394 |
-
with gr.Column(scale=3, min_width=300):
|
395 |
-
input_text_persian = gr.Textbox(
|
396 |
-
lines=4, label="متن فارسی برای ترجمه",
|
397 |
-
placeholder="مثال: سلام، فردا هوا چطور است؟",
|
398 |
-
elem_id="text_input_main"
|
399 |
-
)
|
400 |
-
|
401 |
-
language_dropdown_tts_english = gr.Dropdown(
|
402 |
-
choices=list(language_dict_persian_keys.keys()),
|
403 |
-
value=default_english_tts_voice,
|
404 |
-
label="انتخاب گوینده و لهجه انگلیسی",
|
405 |
-
elem_id="voice_dropdown",
|
406 |
-
interactive=bool(language_dict_persian_keys)
|
407 |
-
|
408 |
-
with gr.Accordion("⚙️ تنظیمات پیشرفته صدا", open=False):
|
409 |
-
gr.Markdown("<p class='advanced-settings-label'>تنظیمات حرفهای برای کنترل دقیقتر خروجی صدا</p>")
|
410 |
-
with gr.Row():
|
411 |
-
rate_slider = gr.Slider(
|
412 |
-
minimum=-100, maximum=100, step=1, value=0,
|
413 |
-
label="سرعت (%)", scale=1, elem_id="rate_slider")
|
414 |
-
volume_slider = gr.Slider(
|
415 |
-
minimum=-100, maximum=100, step=1, value=0,
|
416 |
-
label="حجم (%)", scale=1, elem_id="volume_slider")
|
417 |
-
pitch_slider = gr.Slider(
|
418 |
-
minimum=-50, maximum=50, step=1, value=0,
|
419 |
-
label="گام (Hz)", elem_id="pitch_slider")
|
420 |
-
|
421 |
-
submit_button = gr.Button(
|
422 |
-
"🚀 ترجمه و تلفظ",
|
423 |
-
variant="primary",
|
424 |
-
elem_classes=["generate-button-final"])
|
425 |
-
|
426 |
-
with gr.Column(scale=2, min_width=280):
|
427 |
-
output_text_translated = gr.Textbox(
|
428 |
-
label="متن ترجمه شده (انگلیسی)",
|
429 |
-
interactive=False, lines=6,
|
430 |
-
placeholder="متن انگلیسی ترجمه شده یا پیامهای خطا...",
|
431 |
-
elem_id="translated_text_output")
|
432 |
-
|
433 |
-
output_audio = gr.Audio(
|
434 |
-
type="filepath", label="فایل صوتی",
|
435 |
-
format="mp3", interactive=False,
|
436 |
-
autoplay=True, elem_id="output_audio_player")
|
437 |
-
|
438 |
-
if language_dict_persian_keys and default_english_tts_voice and default_english_tts_voice != "لیست صداها خالی است":
|
439 |
-
gr.HTML("<hr style='height:1px;background-color:var(--app-border-color);margin:1.5rem 0;border:none;'>")
|
440 |
|
441 |
-
|
442 |
-
|
443 |
-
|
444 |
-
|
445 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
446 |
|
447 |
-
|
448 |
-
|
449 |
-
|
450 |
-
["ببخشید، متوجه نشدم. امکان دارد تکرار کنید؟", voice_keys[voice3_idx] if num_voices > 0 else current_default_voice, -10, 0, 0],
|
451 |
-
]
|
452 |
|
453 |
-
|
454 |
-
|
455 |
-
|
456 |
-
|
457 |
-
|
458 |
-
|
459 |
-
cache_examples=os.getenv("GRADIO_CACHE_EXAMPLES", "False").lower() == "true",
|
460 |
-
label="💡 نمونههای کاربردی"
|
461 |
-
)
|
462 |
-
else:
|
463 |
-
gr.Markdown("<p style='text-align:center; color:var(--app-text-secondary); margin-top:1rem;'>نمونهای برای نمایش موجود نیست.</p>")
|
464 |
|
465 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
466 |
|
467 |
-
|
468 |
-
|
469 |
-
inputs=[input_text_persian, language_dropdown_tts_english, rate_slider, volume_slider, pitch_slider],
|
470 |
-
outputs=[output_text_translated, output_audio]
|
471 |
-
)
|
472 |
|
473 |
-
# --- راهاندازی برنامه ---
|
474 |
-
if __name__ == "__main__":
|
475 |
-
if not language_dict_persian_keys:
|
476 |
-
logging.critical("خطای بحرانی: language_dict_persian_keys خالی است!")
|
477 |
|
478 |
-
#
|
479 |
-
|
480 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
481 |
|
482 |
demo.launch(
|
483 |
server_name="0.0.0.0",
|
|
|
1 |
import gradio as gr
|
2 |
+
# import base64 # Not used in your original core logic
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
import mimetypes
|
4 |
+
import os
|
5 |
import re
|
6 |
import struct
|
7 |
+
import time
|
8 |
+
# import zipfile # Not used in your original core logic
|
9 |
+
from google import genai
|
10 |
+
from google.genai import types as genai_types # Aliased your 'types' import to avoid conflict
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
|
12 |
+
# --- Standard Python Logging (very basic) ---
|
13 |
+
import logging
|
14 |
+
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
|
15 |
+
|
16 |
+
try:
|
17 |
+
from pydub import AudioSegment
|
18 |
+
PYDUB_AVAILABLE = True
|
19 |
+
except ImportError:
|
20 |
+
PYDUB_AVAILABLE = False
|
21 |
+
logging.warning("Pydub (for audio merging) not found. Merging will be disabled if multiple audio chunks are generated.")
|
22 |
+
|
23 |
+
|
24 |
+
# --- START: YOUR EXACT CORE TTS LOGIC (AlphaTTS_Original) ---
|
25 |
+
# This entire block is copied VERBATIM from the AlphaTTS_Original code you provided
|
26 |
+
# No changes to logic, variable names, or function signatures within this block.
|
27 |
+
|
28 |
+
SPEAKER_VOICES = [
|
29 |
+
"Achird", "Zubenelgenubi", "Vindemiatrix", "Sadachbia", "Sadaltager",
|
30 |
+
"Sulafat", "Laomedeia", "Achernar", "Alnilam", "Schedar", "Gacrux",
|
31 |
+
"Pulcherrima", "Umbriel", "Algieba", "Despina", "Erinome", "Algenib",
|
32 |
+
"Rasalthgeti", "Orus", "Aoede", "Callirrhoe", "Autonoe", "Enceladus",
|
33 |
+
"Iapetus", "Zephyr", "Puck", "Charon", "Kore", "Fenrir", "Leda"
|
34 |
+
]
|
35 |
+
FIXED_MODEL_NAME = "gemini-2.5-flash-preview-tts" # YOUR DEFINED MODEL
|
36 |
+
DEFAULT_MAX_CHUNK_SIZE = 3800
|
37 |
+
DEFAULT_SLEEP_BETWEEN_REQUESTS = 8
|
38 |
+
DEFAULT_OUTPUT_FILENAME_BASE = "alpha_tts_audio"
|
39 |
+
|
40 |
+
def _log(message, log_list): # YOUR _log function
|
41 |
+
log_list.append(message)
|
42 |
+
logging.info(f"[AlphaTTS_LOG_INTERNAL] {message}") # Added standard logging for visibility
|
43 |
+
|
44 |
+
def save_binary_file(file_name, data, log_list): # YOUR save_binary_file
|
45 |
+
try:
|
46 |
+
with open(file_name, "wb") as f: f.write(data)
|
47 |
+
_log(f"✅ فایل ذخیره شد: {file_name}", log_list)
|
48 |
+
return file_name
|
49 |
+
except Exception as e:
|
50 |
+
_log(f"❌ خطا در ذخیره فایل {file_name}: {e}", log_list)
|
51 |
+
return None
|
52 |
+
|
53 |
+
def convert_to_wav(audio_data: bytes, mime_type: str) -> bytes: # YOUR convert_to_wav
|
54 |
+
parameters = parse_audio_mime_type(mime_type)
|
55 |
+
bits_per_sample, rate = parameters["bits_per_sample"], parameters["rate"]
|
56 |
+
num_channels, data_size = 1, len(audio_data)
|
57 |
+
bytes_per_sample, block_align = bits_per_sample // 8, num_channels * (bits_per_sample // 8)
|
58 |
+
byte_rate, chunk_size = rate * block_align, 36 + data_size
|
59 |
+
header = struct.pack("<4sI4s4sIHHIIHH4sI", b"RIFF", chunk_size, b"WAVE", b"fmt ", 16, 1, num_channels, rate, byte_rate, block_align, bits_per_sample, b"data", data_size)
|
60 |
+
return header + audio_data
|
61 |
+
|
62 |
+
def parse_audio_mime_type(mime_type: str) -> dict[str, int]: # YOUR parse_audio_mime_type
|
63 |
+
bits, rate = 16, 24000
|
64 |
+
for param in mime_type.split(";"):
|
65 |
+
param = param.strip()
|
66 |
+
if param.lower().startswith("rate="):
|
67 |
+
try: rate = int(param.split("=", 1)[1])
|
68 |
+
except ValueError: pass # Your original had 'except: pass'
|
69 |
+
elif param.startswith("audio/L"):
|
70 |
+
try: bits = int(param.split("L", 1)[1])
|
71 |
+
except ValueError: pass # Your original had 'except: pass'
|
72 |
+
return {"bits_per_sample": bits, "rate": rate}
|
73 |
+
|
74 |
+
def smart_text_split(text, max_size=3800, log_list=None): # YOUR smart_text_split
|
75 |
+
if len(text) <= max_size: return [text]
|
76 |
+
chunks, current_chunk = [], ""
|
77 |
+
sentences = re.split(r'(?<=[.!?؟])\s+', text) # Your original regex
|
78 |
+
for sentence in sentences:
|
79 |
+
if len(current_chunk) + len(sentence) + 1 > max_size:
|
80 |
+
if current_chunk: chunks.append(current_chunk.strip())
|
81 |
+
current_chunk = sentence
|
82 |
+
while len(current_chunk) > max_size:
|
83 |
+
split_idx = next((i for i in range(max_size - 1, max_size // 2, -1) if current_chunk[i] in ['،', ',', ';', ':', ' ']), -1)
|
84 |
+
part, current_chunk = (current_chunk[:split_idx+1], current_chunk[split_idx+1:]) if split_idx != -1 else (current_chunk[:max_size], current_chunk[max_size:])
|
85 |
+
chunks.append(part.strip())
|
86 |
+
else: current_chunk += (" " if current_chunk else "") + sentence
|
87 |
+
if current_chunk: chunks.append(current_chunk.strip())
|
88 |
+
final_chunks = [c for c in chunks if c]
|
89 |
+
if log_list: _log(f"📊 متن به {len(final_chunks)} قطعه تقسیم شد.", log_list)
|
90 |
+
return final_chunks
|
91 |
+
|
92 |
+
def merge_audio_files_func(file_paths, output_path, log_list): # YOUR merge_audio_files_func
|
93 |
+
if not PYDUB_AVAILABLE: _log("❌ pydub در دسترس نیست.", log_list); return False
|
94 |
+
try:
|
95 |
+
_log(f"🔗 ادغام {len(file_paths)} فایل صوتی...", log_list)
|
96 |
+
combined = AudioSegment.empty()
|
97 |
+
for i, fp in enumerate(file_paths):
|
98 |
+
if os.path.exists(fp):
|
99 |
+
try:
|
100 |
+
segment = AudioSegment.from_file(fp) # Pydub infers format
|
101 |
+
combined += segment
|
102 |
+
if i < len(file_paths) - 1:
|
103 |
+
combined += AudioSegment.silent(duration=150) # Your original logic
|
104 |
+
except Exception as e_pydub: # More specific exception handling for pydub
|
105 |
+
_log(f"⚠️ خطای Pydub در پردازش فایل '{fp}': {e_pydub}. این فایل نادیده گرفته می شود.", log_list)
|
106 |
+
continue
|
107 |
+
else: _log(f"⚠️ فایل پیدا نشد: {fp}", log_list)
|
108 |
+
if len(combined) == 0: # Check if anything was combined
|
109 |
+
_log("❌ هیچ قطعه صوتی معتبری برای ادغام یافت نشد.", log_list)
|
110 |
+
return False
|
111 |
+
combined.export(output_path, format="wav")
|
112 |
+
_log(f"✅ فایل ادغام شده: {output_path}", log_list); return True
|
113 |
+
except Exception as e: _log(f"❌ خطا در ادغام: {e}", log_list); return False
|
114 |
+
|
115 |
+
def core_generate_audio(text_input, prompt_input, selected_voice, temperature_val, log_list): # YOUR core_generate_audio
|
116 |
+
output_base_name = DEFAULT_OUTPUT_FILENAME_BASE
|
117 |
+
max_chunk, sleep_time = DEFAULT_MAX_CHUNK_SIZE, DEFAULT_SLEEP_BETWEEN_REQUESTS
|
118 |
+
_log(f"🚀 شروع فرآیند با مدل: {FIXED_MODEL_NAME}...", log_list)
|
119 |
+
api_key = os.environ.get("GEMINI_API_KEY") # YOUR API KEY METHOD
|
120 |
+
if not api_key: _log("❌ کلید API تنظیم نشده.", log_list); return None
|
121 |
+
try: client = genai.Client(api_key=api_key) # YOUR CLIENT INSTANTIATION
|
122 |
+
except Exception as e: _log(f"❌ خطا در کلاینت: {e}", log_list); return None # Your original error handling
|
123 |
+
if not text_input or not text_input.strip(): _log("❌ متن ورودی خالی.", log_list); return None
|
124 |
+
text_chunks = smart_text_split(text_input, max_chunk, log_list)
|
125 |
+
if not text_chunks: _log("❌ متن قابل پردازش نیست.", log_list); return None
|
126 |
|
127 |
+
generated_files = []
|
128 |
+
for i, chunk in enumerate(text_chunks):
|
129 |
+
_log(f"🔊 پردازش قطعه {i+1}/{len(text_chunks)} (صدا: {selected_voice}, دما: {temperature_val})...", log_list)
|
130 |
+
final_text = f'"{prompt_input}"\n{chunk}' if prompt_input and prompt_input.strip() else chunk
|
131 |
+
contents = [genai_types.Content(role="user", parts=[genai_types.Part.from_text(text=final_text)])]
|
132 |
+
config = genai_types.GenerateContentConfig(temperature=temperature_val, response_modalities=["audio"],
|
133 |
+
speech_config=genai_types.SpeechConfig(voice_config=genai_types.VoiceConfig(
|
134 |
+
prebuilt_voice_config=genai_types.PrebuiltVoiceConfig(voice_name=selected_voice))))
|
135 |
+
fname_base = f"{output_base_name}_part{i+1:03d}"
|
136 |
try:
|
137 |
+
# YOUR API CALL
|
138 |
+
response = client.models.generate_content(model=FIXED_MODEL_NAME, contents=contents, config=config)
|
139 |
+
# YOUR RESPONSE HANDLING
|
140 |
+
if response.candidates and response.candidates[0].content and response.candidates[0].content.parts and response.candidates[0].content.parts[0].inline_data:
|
141 |
+
inline_data = response.candidates[0].content.parts[0].inline_data
|
142 |
+
data_buffer = inline_data.data
|
143 |
+
mime_type = inline_data.mime_type
|
144 |
+
_log(f"داده صوتی در candidate.part[0].inline_data برای قطعه {i+1} یافت شد. MIME: {mime_type}", log_list)
|
145 |
+
ext = mimetypes.guess_extension(mime_type) or ".wav"
|
146 |
+
if "audio/L" in mime_type and ext == ".wav":
|
147 |
+
_log(f"تبدیل صدای خام PCM (MIME: {mime_type}) به WAV برای قطعه {i+1}.", log_list)
|
148 |
+
data_buffer = convert_to_wav(data_buffer, mime_type)
|
149 |
+
if not ext.startswith("."): ext = "." + ext
|
150 |
+
fpath = save_binary_file(f"{fname_base}{ext}", data_buffer, log_list)
|
151 |
+
if fpath: generated_files.append(fpath)
|
152 |
+
else: _log(f"⚠️ پاسخ API برای قطعه {i+1} بدون داده صوتی.", log_list) # Your original log message
|
153 |
+
except Exception as e: _log(f"❌ خطا در تولید قطعه {i+1}: {e}", log_list); continue # YOUR ORIGINAL ERROR HANDLING
|
154 |
+
if i < len(text_chunks) - 1 and len(text_chunks) > 1:
|
155 |
+
_log(f"💤 توقف کوتاه ({sleep_time} ثانیه) قبل از قطعه بعدی...", log_list)
|
156 |
+
time.sleep(sleep_time)
|
157 |
+
|
158 |
+
if not generated_files: _log("❌ هیچ فایلی تولید نشد.", log_list); return None
|
159 |
+
_log(f"🎉 {len(generated_files)} فایل(های) صوتی تولید شد.", log_list)
|
160 |
|
161 |
+
final_audio_file = None
|
162 |
+
final_output_path_base = f"{output_base_name}_final"
|
163 |
|
164 |
+
# YOUR FILE MERGING AND RENAMING LOGIC (VERBATIM)
|
165 |
+
if len(generated_files) > 1:
|
166 |
+
if PYDUB_AVAILABLE:
|
167 |
+
merged_fn = f"{final_output_path_base}.wav"
|
168 |
+
if os.path.exists(merged_fn):
|
169 |
+
try: os.remove(merged_fn)
|
170 |
+
except Exception as e_rm: _log(f"⚠️ خطا در حذف فایل ادغام شده قبلی '{merged_fn}': {e_rm}", log_list) # Added log
|
171 |
+
if merge_audio_files_func(generated_files, merged_fn, log_list):
|
172 |
+
final_audio_file = merged_fn
|
173 |
+
for fp in generated_files:
|
174 |
+
if os.path.abspath(fp) != os.path.abspath(merged_fn):
|
175 |
+
try: os.remove(fp)
|
176 |
+
except: pass # Your original silent pass
|
177 |
+
else:
|
178 |
+
if generated_files:
|
179 |
+
try:
|
180 |
+
# Ensuring target name doesn't overwrite source if paths are somehow the same initially
|
181 |
+
source_path = generated_files[0]
|
182 |
+
target_path = f"{final_output_path_base}{os.path.splitext(source_path)[1]}"
|
183 |
+
if os.path.abspath(source_path) != os.path.abspath(target_path):
|
184 |
+
if os.path.exists(target_path): os.remove(target_path)
|
185 |
+
os.rename(source_path, target_path)
|
186 |
+
final_audio_file = target_path
|
187 |
+
# Clean up other parts only if the first one was successfully moved/renamed
|
188 |
+
if final_audio_file == target_path:
|
189 |
+
for i_gf in range(1, len(generated_files)):
|
190 |
+
try: os.remove(generated_files[i_gf])
|
191 |
+
except: pass
|
192 |
+
except Exception as e_rename: # Your original variable name
|
193 |
+
_log(f"خطا در تغییر نام فایل اولین قطعه: {e_rename}", log_list)
|
194 |
+
final_audio_file = generated_files[0]
|
195 |
+
else:
|
196 |
+
_log("⚠️ pydub نیست. اولین قطعه ارائه میشود.", log_list)
|
197 |
+
if generated_files:
|
198 |
+
try:
|
199 |
+
source_path = generated_files[0]
|
200 |
+
target_path = f"{final_output_path_base}{os.path.splitext(source_path)[1]}"
|
201 |
+
if os.path.abspath(source_path) != os.path.abspath(target_path):
|
202 |
+
if os.path.exists(target_path): os.remove(target_path)
|
203 |
+
os.rename(source_path, target_path)
|
204 |
+
final_audio_file = target_path
|
205 |
+
if final_audio_file == target_path:
|
206 |
+
for i_gf in range(1, len(generated_files)):
|
207 |
+
try: os.remove(generated_files[i_gf])
|
208 |
+
except: pass
|
209 |
+
except Exception as e_rename_single: # Your original variable name
|
210 |
+
_log(f"خطا در تغییر نام فایل اولین قطعه (بدون pydub): {e_rename_single}", log_list)
|
211 |
+
final_audio_file = generated_files[0]
|
212 |
+
elif len(generated_files) == 1:
|
213 |
+
try:
|
214 |
+
source_path = generated_files[0]
|
215 |
+
target_path = f"{final_output_path_base}{os.path.splitext(source_path)[1]}"
|
216 |
+
if os.path.abspath(source_path) != os.path.abspath(target_path):
|
217 |
+
if os.path.exists(target_path): os.remove(target_path)
|
218 |
+
os.rename(source_path, target_path)
|
219 |
+
final_audio_file = target_path
|
220 |
+
except Exception as e_rename_single_final: # Your original variable name
|
221 |
+
_log(f"خطا در تغییر نام فایل تکی نهایی: {e_rename_single_final}", log_list)
|
222 |
+
final_audio_file = generated_files[0]
|
223 |
+
|
224 |
+
if final_audio_file and not os.path.exists(final_audio_file):
|
225 |
+
_log(f"⚠️ فایل نهایی '{final_audio_file}' وجود ندارد!", log_list)
|
226 |
+
return None
|
227 |
+
return final_audio_file # YOUR ORIGINAL RETURN
|
228 |
|
229 |
+
def gradio_tts_interface(use_file_input, uploaded_file, text_to_speak, speech_prompt, speaker_voice, temperature, progress=gr.Progress(track_tqdm=True)): # YOUR ORIGINAL SIGNATURE
|
230 |
+
logs = []
|
231 |
+
actual_text = ""
|
232 |
+
if use_file_input:
|
233 |
+
if uploaded_file: # In Gradio, uploaded_file is a TemporaryFileWrapper object
|
234 |
+
try:
|
235 |
+
# uploaded_file.name gives the path to the temporary file
|
236 |
+
with open(uploaded_file.name, 'r', encoding='utf-8') as f: actual_text = f.read().strip()
|
237 |
+
if not actual_text: return None
|
238 |
+
except Exception as e: _log(f"❌ خطا خواندن فایل: {e}", logs); return None
|
239 |
+
else: return None # No file provided when checkbox is true
|
240 |
+
else:
|
241 |
+
actual_text = text_to_speak
|
242 |
+
if not actual_text or not actual_text.strip(): return None
|
243 |
|
244 |
+
final_path = core_generate_audio(actual_text, speech_prompt, speaker_voice, temperature, logs)
|
245 |
+
# for log_entry in logs: print(log_entry) # Your original commented-out log printing
|
246 |
+
return final_path # YOUR ORIGINAL RETURN (returns only audio path)
|
247 |
+
# --- END: YOUR EXACT CORE TTS LOGIC (AlphaTTS_Original) ---
|
248 |
|
|
|
|
|
|
|
|
|
|
|
|
|
249 |
|
250 |
+
# --- START: Styling and UI (Applying AlphaTranslator_Styled look to YOUR UI structure) ---
|
251 |
+
# (Using CSS variables from AlphaTranslator_Styled for colors and fonts)
|
252 |
+
FLY_PRIMARY_COLOR_HEX = "#4F46E5"
|
253 |
+
FLY_SECONDARY_COLOR_HEX = "#10B981"
|
254 |
+
FLY_ACCENT_COLOR_HEX = "#D97706" # Orange accent from AlphaTranslator_Styled (for buttons)
|
255 |
+
FLY_TEXT_COLOR_HEX = "#1F2937"
|
256 |
+
FLY_SUBTLE_TEXT_HEX = "#6B7280"
|
257 |
+
FLY_LIGHT_BACKGROUND_HEX = "#F9FAFB" # Overall page bg
|
258 |
+
FLY_WHITE_HEX = "#FFFFFF" # Panel bg
|
259 |
+
FLY_BORDER_COLOR_HEX = "#D1D5DB"
|
260 |
+
FLY_INPUT_BG_HEX_SIMPLE = "#F3F4F6" # Input bg
|
261 |
+
|
262 |
+
# Theme for Gradio Blocks (using a base font from AlphaTranslator_Styled)
|
263 |
+
# Your original `gr.Blocks` used `theme=gr.themes.Base(font=[gr.themes.GoogleFont("Vazirmatn")])`
|
264 |
+
# We'll use `app_theme_applied_styled` for body background and apply Vazirmatn in CSS.
|
265 |
+
app_theme_applied_styled = gr.themes.Base(
|
266 |
+
font=[gr.themes.GoogleFont("Inter"), "system-ui", "sans-serif"],
|
267 |
+
).set(
|
268 |
+
body_background_fill=FLY_LIGHT_BACKGROUND_HEX,
|
269 |
+
)
|
270 |
+
|
271 |
+
# Combined and adapted CSS
|
272 |
+
# Goal: Make YOUR UI components look like the AlphaTranslator_Styled components.
|
273 |
+
# Your original component IDs are like "use_file_cb_alpha_v3", "generate_button_alpha_v3", etc.
|
274 |
+
# CSS will target these IDs where possible.
|
275 |
+
final_combined_css_v2 = f"""
|
276 |
+
@import url('https://fonts.googleapis.com/css2?family=Vazirmatn:wght@300;400;500;600;700;800&display=swap');
|
277 |
@import url('https://fonts.googleapis.com/css2?family=Poppins:wght@400;500;600;700;800&display=swap');
|
278 |
+
@import url('https://fonts.googleapis.com/css2?family=Inter:wght@400;500;600;700;800&display=swap');
|
279 |
+
|
280 |
:root {{
|
281 |
+
--fly-primary: {FLY_PRIMARY_COLOR_HEX};
|
282 |
+
--fly-secondary: {FLY_SECONDARY_COLOR_HEX};
|
283 |
+
--fly-accent: {FLY_ACCENT_COLOR_HEX};
|
284 |
+
--fly-text-primary: {FLY_TEXT_COLOR_HEX};
|
285 |
+
--fly-text-secondary: {FLY_SUBTLE_TEXT_HEX};
|
286 |
+
--fly-bg-light: {FLY_LIGHT_BACKGROUND_HEX};
|
287 |
+
--fly-bg-white: {FLY_WHITE_HEX};
|
288 |
+
--fly-border-color: {FLY_BORDER_COLOR_HEX};
|
289 |
+
--fly-input-bg-simple: {FLY_INPUT_BG_HEX_SIMPLE};
|
290 |
+
--fly-primary-rgb: 79,70,229;
|
291 |
+
--fly-accent-rgb: 217,119,6;
|
292 |
+
|
293 |
+
--radius-sm: 0.375rem; --radius-md: 0.5rem; --radius-lg: 0.75rem; --radius-xl: 1rem;
|
294 |
+
--shadow-lg: 0 10px 15px -3px rgba(0,0,0,0.1),0 4px 6px -4px rgba(0,0,0,0.1);
|
295 |
+
--shadow-xl: 0 20px 25px -5px rgba(0,0,0,0.1),0 8px 10px -6px rgba(0,0,0,0.1);
|
296 |
+
|
297 |
+
--font-global: 'Vazirmatn', 'Inter', 'Poppins', system-ui, sans-serif;
|
298 |
+
--font-english: 'Poppins', 'Inter', system-ui, sans-serif;
|
299 |
+
|
300 |
+
/* Your original AlphaTTS CSS variables if different and needed for exact match */
|
301 |
+
--app-button-bg-original: #2979FF; /* Blue from your original AlphaTTS button */
|
302 |
+
--radius-input-original: 8px; /* Your original input radius */
|
303 |
}}
|
304 |
+
|
305 |
+
body {{
|
306 |
+
font-family: var(--font-global);
|
307 |
direction: rtl;
|
308 |
+
background-color: var(--fly-bg-light) !important; /* Ensure override */
|
309 |
+
color: var(--fly-text-primary);
|
310 |
+
line-height: 1.7;
|
311 |
+
font-size: 16px;
|
312 |
}}
|
313 |
+
|
314 |
+
.gradio-container {{
|
315 |
+
max-width:100% !important; width:100% !important; min-height:100vh;
|
316 |
+
margin:0 auto !important; padding:0 !important; border-radius:0 !important;
|
317 |
+
box-shadow:none !important; background:linear-gradient(170deg, #E0F2FE 0%, #F3E8FF 100%) !important; /* Ensure override */
|
318 |
+
display:flex; flex-direction:column;
|
|
|
319 |
}}
|
320 |
+
|
321 |
+
/* Header styling from AlphaTranslator_Styled */
|
322 |
+
.app-header-alphatts-v2 {{ /* Unique class for this app's header */
|
323 |
+
text-align:center; padding:2.5rem 1rem; margin:0;
|
324 |
+
background:linear-gradient(135deg, var(--fly-primary) 0%, var(--fly-secondary) 100%);
|
325 |
+
color:var(--fly-bg-white); border-bottom-left-radius:var(--radius-xl);
|
326 |
+
border-bottom-right-radius:var(--radius-xl); box-shadow:var(--shadow-lg);
|
327 |
+
position:relative; overflow:hidden;
|
328 |
}}
|
329 |
+
.app-header-alphatts-v2::before {{
|
330 |
+
content:''; position:absolute; top:-50px; right:-50px; width:150px; height:150px;
|
331 |
+
background:rgba(255,255,255,0.1); border-radius:9999px;
|
332 |
+
opacity:0.5; transform:rotate(45deg);
|
|
|
333 |
}}
|
334 |
+
.app-header-alphatts-v2 h1 {{
|
335 |
+
font-size:2.25em !important; font-weight:800 !important; margin:0 0 0.5rem 0;
|
336 |
+
font-family:var(--font-english); letter-spacing:-0.5px; text-shadow:0 2px 4px rgba(0,0,0,0.1);
|
|
|
|
|
337 |
}}
|
338 |
+
.app-header-alphatts-v2 p {{
|
339 |
+
font-size:1em !important; margin-top:0.25rem; font-weight:400;
|
340 |
+
color:rgba(255,255,255,0.85) !important;
|
341 |
+
}}
|
342 |
+
|
343 |
+
/* Main content panel structure from AlphaTranslator_Styled */
|
344 |
+
.main-content-area-alphatts-v2 {{
|
345 |
+
flex-grow:1; padding:0.75rem; width:100%; margin:0 auto; box-sizing:border-box;
|
|
|
|
|
346 |
}}
|
347 |
+
/* This is the direct parent of your UI elements if you keep gr.Column as the main wrapper */
|
348 |
+
.content-panel-alphatts-v2 {{
|
349 |
+
background-color:var(--fly-bg-white); padding:1rem; border-radius:var(--radius-xl);
|
350 |
+
box-shadow:var(--shadow-xl); margin-top:-2rem; position:relative; z-index:10;
|
351 |
+
margin-bottom:2rem; width:100%; box-sizing:border-box;
|
352 |
+
}}
|
353 |
+
|
354 |
+
/* Styling YOUR UI elements to match AlphaTranslator_Styled look */
|
355 |
+
/* Inputs: Targeting specific elem_ids from your code where possible */
|
356 |
+
.content-panel-alphatts-v2 #{generate_button_alpha_v3} + .gr-input > label + div > textarea, /* For text_to_speak_tb */
|
357 |
+
.content-panel-alphatts-v2 #{generate_button_alpha_v3} + .gr-dropdown > label + div > div > input,
|
358 |
+
.content-panel-alphatts-v2 #{generate_button_alpha_v3} + .gr-dropdown > label + div > div > select,
|
359 |
+
.content-panel-alphatts-v2 #{generate_button_alpha_v3} + .gr-textbox > label + div > textarea, /* For speech_prompt_tb */
|
360 |
+
.content-panel-alphatts-v2 #{generate_button_alpha_v3} + .gr-file > label + div, /* For uploaded_file_input */
|
361 |
+
.content-panel-alphatts-v2 .gr-slider /* General slider */
|
362 |
+
{{
|
363 |
+
border-radius:var(--radius-input-original) !important; /* YOUR original radius */
|
364 |
+
border:1.5px solid var(--fly-border-color) !important;
|
365 |
+
font-size:0.95em !important;
|
366 |
+
background-color:var(--fly-input-bg-simple) !important;
|
367 |
+
padding:10px 12px !important;
|
368 |
+
color:var(--fly-text-primary) !important;
|
369 |
+
box-shadow: none !important;
|
370 |
+
}}
|
371 |
+
|
372 |
+
/* Focus styles for inputs */
|
373 |
+
.content-panel-alphatts-v2 .gr-input > label + div > textarea:focus,
|
374 |
+
.content-panel-alphatts-v2 .gr-dropdown > label + div > div > input:focus,
|
375 |
+
.content-panel-alphatts-v2 .gr-dropdown > label + div > div > select:focus,
|
376 |
+
.content-panel-alphatts-v2 .gr-textbox > label + div > textarea:focus,
|
377 |
+
.content-panel-alphatts-v2 .gr-file > label + div:focus-within
|
378 |
+
{{
|
379 |
+
border-color:var(--fly-primary) !important;
|
380 |
+
box-shadow:0 0 0 3px rgba(var(--fly-primary-rgb),0.12) !important;
|
381 |
+
background-color:var(--fly-bg-white) !important;
|
382 |
}}
|
383 |
+
.content-panel-alphatts-v2 .gr-file > label + div {{ text-align:center; border-style: dashed !important; }}
|
384 |
+
|
385 |
+
/* Button: Targeting YOUR button by its elem_id with AlphaTranslator_Styled primary button look */
|
386 |
+
.content-panel-alphatts-v2 .gr-button#{generate_button_alpha_v3} /* YOUR BUTTON ID */
|
387 |
+
{{
|
388 |
+
background:var(--fly-accent) !important; /* Orange accent */
|
389 |
+
margin-top:1.5rem !important; padding:12px 20px !important; /* Adjusted padding */
|
390 |
+
transition:all 0.25s ease-in-out !important; color:white !important; font-weight:600 !important;
|
391 |
+
border-radius:var(--radius-input-original) !important; /* Your original radius */ border:none !important;
|
392 |
+
box-shadow:0 3px 8px -1px rgba(var(--fly-accent-rgb),0.3) !important;
|
393 |
+
width:100% !important; font-size:1.05em !important; /* Your original font size */
|
394 |
+
display:flex; align-items:center; justify-content:center;
|
|
|
|
|
395 |
}}
|
396 |
+
.content-panel-alphatts-v2 .gr-button#{generate_button_alpha_v3}:hover
|
397 |
+
{{
|
398 |
+
background:#B45309 !important; transform:translateY(-1px) !important;
|
399 |
+
box-shadow:0 5px 10px -1px rgba(var(--fly-accent-rgb),0.4) !important;
|
400 |
}}
|
401 |
+
|
402 |
+
/* Labels: General style (less specific to avoid breaking Gradio's structure) */
|
403 |
+
.content-panel-alphatts-v2 .gr-form label > .label-text span /* Try to target the span inside label */
|
404 |
+
{{
|
405 |
+
font-weight:500 !important; color: var(--fly-text-secondary) !important;
|
406 |
+
font-size:0.88em !important; margin-bottom:6px !important; display:inline-block;
|
|
|
|
|
|
|
407 |
}}
|
408 |
+
/* Your temperature description class */
|
409 |
+
.content-panel-alphatts-v2 .temp_description_class_alpha_v3 {{
|
410 |
+
font-size: 0.85em; color: var(--fly-text-secondary); margin-top: -0.4rem; margin-bottom: 1rem;
|
|
|
|
|
411 |
}}
|
412 |
+
|
413 |
+
/* Audio Player: Targeting your specific elem_id */
|
414 |
+
.content-panel-alphatts-v2 #output_audio_player_alpha_v3 audio
|
415 |
+
{{
|
416 |
+
width: 100%; border-radius: var(--radius-input-original); margin-top:0.8rem;
|
417 |
}}
|
418 |
+
|
419 |
+
/* Examples styling from AlphaTranslator_Styled */
|
420 |
+
.content-panel-alphatts-v2 .gr-examples .gr-button.gr-button-tool,
|
421 |
+
.content-panel-alphatts-v2 .gr-examples .gr-sample-button
|
422 |
+
{{
|
423 |
+
background-color:#E0E7FF !important; color:var(--fly-primary) !important;
|
424 |
+
border-radius:var(--radius-sm) !important; font-size:0.78em !important; padding:4px 8px !important;
|
425 |
}}
|
426 |
+
.content-panel-alphatts-v2 .custom-hr {{height:1px;background-color:var(--fly-border-color);margin:1.5rem 0;border:none;}}
|
427 |
+
|
428 |
+
/* Footer styling from AlphaTranslator_Styled */
|
429 |
+
.app-footer-alphatts-v2 {{
|
430 |
+
text-align:center;font-size:0.85em;color:var(--fly-text-secondary);margin-top:2.5rem;
|
431 |
+
padding:1rem 0;background-color:rgba(255,255,255,0.3);backdrop-filter:blur(5px);
|
432 |
+
border-top:1px solid var(--fly-border-color);
|
|
|
433 |
}}
|
434 |
+
/* Hide default Gradio watermarks/footers more aggressively */
|
435 |
+
footer, .gradio-footer, .flagging-container, .footer-utils, .footer, div[class*="svelte-"], .spacer.svelte-164SbmI {{
|
436 |
+
display:none !important; visibility:hidden !important; opacity: 0 !important; height: 0 !important;
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
437 |
}}
|
438 |
+
|
439 |
+
|
440 |
+
/* Responsive adjustments from AlphaTranslator_Styled */
|
441 |
+
@media (min-width:640px) {{
|
442 |
+
.main-content-area-alphatts-v2 {{padding:1.5rem;max-width:700px;}}
|
443 |
+
.content-panel-alphatts-v2 {{padding:1.5rem;}}
|
444 |
+
.app-header-alphatts-v2 h1 {{font-size:2.5em !important;}}
|
445 |
+
.app-header-alphatts-v2 p {{font-size:1.05em !important;}}
|
446 |
}}
|
447 |
+
@media (min-width:768px) {{
|
448 |
+
.main-content-area-alphatts-v2 {{max-width:780px;}}
|
449 |
+
.content-panel-alphatts-v2 {{padding:2rem;}}
|
450 |
+
.content-panel-alphatts-v2 .gr-button#{generate_button_alpha_v3} /* Your button ID */
|
451 |
+
{{
|
452 |
+
width:auto !important; align-self:flex-start;
|
453 |
+
}}
|
454 |
+
.app-header-alphatts-v2 h1 {{font-size:2.75em !important;}}
|
455 |
+
.app-header-alphatts-v2 p {{font-size:1.1em !important;}}
|
456 |
}}
|
457 |
"""
|
458 |
|
459 |
+
# --- Gradio UI Definition (YOUR UI structure, with new CSS applied) ---
|
460 |
+
# Using app_theme_applied_styled for base theme, and final_combined_css_v2 for specifics
|
461 |
+
# The theme for gr.Blocks is `gr.themes.Base(font=[gr.themes.GoogleFont("Vazirmatn")])` from YOUR original code.
|
462 |
+
# We will use `app_theme_applied_styled` for the overall page, and `final_combined_css_v2` for detailed styling.
|
463 |
+
with gr.Blocks(theme=app_theme_applied_styled, css=final_combined_css_v2, title=f"آلفا TTS ({FIXED_MODEL_NAME.split('-')[1]})") as demo:
|
464 |
+
# Header from AlphaTranslator_Styled structure, with a unique class name
|
465 |
+
gr.HTML(f"""
|
466 |
+
<div class='app-header-alphatts-v2'>
|
467 |
+
<h1>🚀 Alpha TTS</h1>
|
468 |
+
<p>جادوی تبدیل متن به صدا در دستان شما (Gemini {FIXED_MODEL_NAME.split('-')[1]})</p>
|
469 |
+
</div>
|
|
|
|
|
|
|
|
|
470 |
""")
|
471 |
|
472 |
+
# Main content area wrapper from AlphaTranslator_Styled structure
|
473 |
+
with gr.Column(elem_classes=["main-content-area-alphatts-v2"]):
|
474 |
+
# Content panel wrapper from AlphaTranslator_Styled structure
|
475 |
+
# This will wrap YOUR gr.Column that contains all UI elements.
|
476 |
+
with gr.Column(elem_classes=["content-panel-alphatts-v2"]):
|
477 |
+
|
478 |
+
# YOUR ORIGINAL UI LAYOUT (from your provided AlphaTTS_Original code)
|
479 |
+
# All `elem_id`s are from your original code.
|
480 |
+
|
481 |
+
# Optional: Warning if GEMINI_API_KEY is not set (simple version)
|
482 |
+
if not os.environ.get("GEMINI_API_KEY"):
|
483 |
+
gr.Markdown("<p style='color:red; text-align:center; margin-bottom:1rem;'>⚠️ <b>هشدار:</b> متغیر محیطی GEMINI_API_KEY تنظیم نشده است. برنامه کار نخواهد کرد.</p>")
|
484 |
+
|
485 |
+
use_file_input_cb = gr.Checkbox(label="📄 استفاده از فایل متنی (.txt)", value=False, elem_id="use_file_cb_alpha_v3")
|
486 |
+
|
487 |
+
uploaded_file_input = gr.File(
|
488 |
+
label=" ",
|
489 |
+
file_types=['.txt'],
|
490 |
+
visible=False,
|
491 |
+
elem_id="file_uploader_alpha_main_v3"
|
492 |
+
)
|
493 |
+
|
494 |
+
text_to_speak_tb = gr.Textbox(
|
495 |
+
label="متن فارسی برای تبدیل",
|
496 |
+
placeholder="مثال: سلام، فردا هوا چطور است؟",
|
497 |
+
lines=5,
|
498 |
+
value="",
|
499 |
+
visible=True,
|
500 |
+
elem_id="text_input_main_alpha_v3"
|
501 |
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
502 |
|
503 |
+
# YOUR ORIGINAL change function for checkbox
|
504 |
+
use_file_input_cb.change(
|
505 |
+
fn=lambda x: (gr.update(visible=x, label=" " if x else "متن فارسی برای تبدیل"), gr.update(visible=not x)),
|
506 |
+
inputs=use_file_input_cb,
|
507 |
+
outputs=[uploaded_file_input, text_to_speak_tb]
|
508 |
+
)
|
509 |
+
|
510 |
+
speech_prompt_tb = gr.Textbox(
|
511 |
+
label="سبک گفتار (اختیاری)",
|
512 |
+
placeholder="مثال: با لحنی شاد و پرانرژی",
|
513 |
+
value="با لحنی دوستانه و رسا صحبت کن.",
|
514 |
+
lines=2, elem_id="speech_prompt_alpha_v3"
|
515 |
+
)
|
516 |
|
517 |
+
speaker_voice_dd = gr.Dropdown(
|
518 |
+
SPEAKER_VOICES, label="انتخاب گوینده و لهجه", value="Charon", elem_id="speaker_voice_alpha_v3"
|
519 |
+
)
|
|
|
|
|
520 |
|
521 |
+
temperature_slider = gr.Slider(
|
522 |
+
minimum=0.1, maximum=1.5, step=0.05, value=0.9, label="میزان خلاقیت صدا",
|
523 |
+
elem_id="temperature_slider_alpha_v3"
|
524 |
+
)
|
525 |
+
# Your original temperature description
|
526 |
+
gr.Markdown("<p class='temp_description_class_alpha_v3'>مقادیر بالاتر = تنوع بیشتر، مقادیر پایینتر = یکنواختی بیشتر.</p>")
|
|
|
|
|
|
|
|
|
|
|
527 |
|
528 |
+
# Your original button, CSS will target its elem_id
|
529 |
+
generate_button = gr.Button("🚀 تولید و پخش صدا", elem_id="generate_button_alpha_v3")
|
530 |
+
|
531 |
+
# Your original audio output, CSS will target its elem_id
|
532 |
+
output_audio = gr.Audio(label=" ", type="filepath", elem_id="output_audio_player_alpha_v3")
|
533 |
+
|
534 |
+
# Your original Examples section
|
535 |
+
gr.HTML("<hr class='custom-hr'>") # Using the styled HR
|
536 |
+
gr.Markdown( # Simplified title styling
|
537 |
+
"<h3 style='text-align:center; font-weight:500; color:var(--fly-text-secondary); margin-top:1.5rem; margin-bottom:1rem;'>نمونههای کاربردی</h3>"
|
538 |
+
)
|
539 |
+
gr.Examples(
|
540 |
+
examples=[
|
541 |
+
[False, None, "سلام بر شما، امیدوارم روز خوبی داشته باشید.", "با لحنی گرم و صمیمی.", "Zephyr", 0.85],
|
542 |
+
[False, None, "این یک آزمایش برای بررسی کیفیت صدای تولید شده توسط هوش مصنوعی آلفا است.", "با صدایی طبیعی و روان.", "Charon", 0.9],
|
543 |
+
],
|
544 |
+
inputs=[use_file_input_cb, uploaded_file_input, text_to_speak_tb, speech_prompt_tb, speaker_voice_dd, temperature_slider],
|
545 |
+
outputs=[output_audio], # YOUR ORIGINAL OUTPUTS FOR EXAMPLES
|
546 |
+
fn=gradio_tts_interface,
|
547 |
+
cache_examples=os.getenv("GRADIO_CACHE_EXAMPLES", "False").lower() == "true"
|
548 |
+
)
|
549 |
|
550 |
+
# Footer from AlphaTranslator_Styled structure, with unique class name
|
551 |
+
gr.Markdown(f"<p class='app-footer-alphatts-v2'>Alpha TTS © 2024 - Model: {FIXED_MODEL_NAME.split('-')[0].upper()} {FIXED_MODEL_NAME.split('-')[1]}</p>")
|
|
|
|
|
|
|
552 |
|
|
|
|
|
|
|
|
|
553 |
|
554 |
+
# --- Event Handlers (YOUR ORIGINAL EVENT HANDLERS) ---
|
555 |
+
if generate_button is not None:
|
556 |
+
generate_button.click(
|
557 |
+
fn=gradio_tts_interface,
|
558 |
+
inputs=[use_file_input_cb, uploaded_file_input, text_to_speak_tb, speech_prompt_tb, speaker_voice_dd, temperature_slider, gr.Progress(track_tqdm=True)], # Added progress back as per your original signature
|
559 |
+
outputs=[output_audio]
|
560 |
+
)
|
561 |
+
else:
|
562 |
+
logging.error("دکمه تولید صدا (generate_button_alpha_v3) در UI یافت نشد.")
|
563 |
+
|
564 |
+
|
565 |
+
if __name__ == "__main__":
|
566 |
+
if not PYDUB_AVAILABLE:
|
567 |
+
logging.warning("Pydub (for audio merging) not found. Merging will be disabled if multiple audio chunks are generated.")
|
568 |
+
if not os.environ.get("GEMINI_API_KEY"):
|
569 |
+
logging.warning("GEMINI_API_KEY environment variable not set. TTS functionality WILL FAIL.")
|
570 |
|
571 |
demo.launch(
|
572 |
server_name="0.0.0.0",
|