Spaces:
Configuration error
Configuration error
Update app.py
Browse files
app.py
CHANGED
@@ -0,0 +1,201 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import gradio as gr
|
2 |
+
import whisper
|
3 |
+
import os
|
4 |
+
import asyncio
|
5 |
+
import edge_tts
|
6 |
+
from transformers import pipeline
|
7 |
+
from deep_translator import GoogleTranslator
|
8 |
+
from docx import Document
|
9 |
+
import tempfile
|
10 |
+
from datetime import datetime
|
11 |
+
import logging
|
12 |
+
from pydub import AudioSegment
|
13 |
+
|
14 |
+
# إعداد التسجيل
|
15 |
+
logging.basicConfig(
|
16 |
+
level=logging.INFO,
|
17 |
+
format="%(asctime)s - %(name)s - %(levelname)s - %(message)s",
|
18 |
+
handlers=[
|
19 |
+
logging.FileHandler("app.log"),
|
20 |
+
logging.StreamHandler(),
|
21 |
+
],
|
22 |
+
)
|
23 |
+
logger = logging.getLogger(__name__)
|
24 |
+
|
25 |
+
# قائمة اللغات المدعومة
|
26 |
+
SUPPORTED_LANGUAGES = {
|
27 |
+
"ar": "العربية",
|
28 |
+
"en": "English",
|
29 |
+
"fr": "Français",
|
30 |
+
"es": "Español",
|
31 |
+
"de": "Deutsch",
|
32 |
+
}
|
33 |
+
|
34 |
+
# تعيين أصوات لكل لغة
|
35 |
+
VOICE_MAPPINGS = {
|
36 |
+
"ar": "ar-EG-ShakirNeural",
|
37 |
+
"en": "en-US-EricNeural",
|
38 |
+
"fr": "fr-FR-HenriNeural",
|
39 |
+
"es": "es-ES-AlvaroNeural",
|
40 |
+
"de": "de-DE-ConradNeural",
|
41 |
+
}
|
42 |
+
|
43 |
+
# تحديد اللغات RTL
|
44 |
+
RTL_LANGUAGES = ["ar"]
|
45 |
+
|
46 |
+
# وظيفة تنظيف النص
|
47 |
+
def clean_text(text):
|
48 |
+
return (
|
49 |
+
text.replace("’", "'")
|
50 |
+
.replace("«", "")
|
51 |
+
.replace("»", "")
|
52 |
+
.replace("\n", " ")
|
53 |
+
.strip()
|
54 |
+
)
|
55 |
+
|
56 |
+
# وظيفة توليد الصوت
|
57 |
+
async def generate_speech(text, lang):
|
58 |
+
"""توليد الصوت باستخدام edge-tts"""
|
59 |
+
try:
|
60 |
+
voice = VOICE_MAPPINGS.get(lang, "en-US-EricNeural")
|
61 |
+
communicate = edge_tts.Communicate(text, voice)
|
62 |
+
audio_path = tempfile.mktemp(suffix=".mp3")
|
63 |
+
await communicate.save(audio_path)
|
64 |
+
|
65 |
+
if os.path.exists(audio_path) and os.path.getsize(audio_path) > 0:
|
66 |
+
logger.info(f"تم إنشاء ملف صوتي: {audio_path}")
|
67 |
+
return audio_path
|
68 |
+
else:
|
69 |
+
logger.error("فشل إنشاء ملف صوتي صالح")
|
70 |
+
return None
|
71 |
+
|
72 |
+
except Exception as e:
|
73 |
+
logger.error(f"خطأ في توليد الصوت: {str(e)}")
|
74 |
+
return None
|
75 |
+
|
76 |
+
# تحويل النص إلى صوت
|
77 |
+
def text_to_speech(text, lang):
|
78 |
+
if not text:
|
79 |
+
logger.warning("لم يتم تقديم نص للتحويل إلى صوت")
|
80 |
+
return None
|
81 |
+
|
82 |
+
try:
|
83 |
+
text = clean_text(text)
|
84 |
+
max_length = 1000
|
85 |
+
text_parts = [text[i : i + max_length] for i in range(0, len(text), max_length)]
|
86 |
+
|
87 |
+
audio_files = []
|
88 |
+
for part in text_parts:
|
89 |
+
audio_path = asyncio.run(generate_speech(part, lang))
|
90 |
+
if audio_path:
|
91 |
+
audio_files.append(audio_path)
|
92 |
+
|
93 |
+
if len(audio_files) == 1:
|
94 |
+
return audio_files[0]
|
95 |
+
|
96 |
+
final_audio = AudioSegment.from_mp3(audio_files[0])
|
97 |
+
for audio_file in audio_files[1:]:
|
98 |
+
final_audio += AudioSegment.from_mp3(audio_file)
|
99 |
+
|
100 |
+
final_path = tempfile.mktemp(suffix=".mp3")
|
101 |
+
final_audio.export(final_path, format="mp3")
|
102 |
+
|
103 |
+
for file in audio_files:
|
104 |
+
os.remove(file)
|
105 |
+
|
106 |
+
return final_path
|
107 |
+
|
108 |
+
except Exception as e:
|
109 |
+
logger.error(f"خطأ في تحويل النص إلى صوت: {str(e)}")
|
110 |
+
return None
|
111 |
+
|
112 |
+
# الترجمة
|
113 |
+
def translate_text(text, source_lang, target_lang):
|
114 |
+
if source_lang == target_lang:
|
115 |
+
return text
|
116 |
+
|
117 |
+
try:
|
118 |
+
translator = GoogleTranslator(source=source_lang, target=target_lang)
|
119 |
+
max_length = 5000
|
120 |
+
text_parts = [text[i : i + max_length] for i in range(0, len(text), max_length)]
|
121 |
+
translated_parts = [translator.translate(part) for part in text_parts]
|
122 |
+
return " ".join(translated_parts)
|
123 |
+
|
124 |
+
except Exception as e:
|
125 |
+
logger.error(f"خطأ في الترجمة: {str(e)}")
|
126 |
+
return f"خطأ في الترجمة: {str(e)}"
|
127 |
+
|
128 |
+
# معالجة الفيديو
|
129 |
+
def process_video(video, source_lang="en", target_lang="ar"):
|
130 |
+
if video is None:
|
131 |
+
return {
|
132 |
+
"error": "الرجاء رفع ملف فيديو",
|
133 |
+
"original": "",
|
134 |
+
"translated": "",
|
135 |
+
}
|
136 |
+
|
137 |
+
try:
|
138 |
+
temp_path = video.name
|
139 |
+
model = whisper.load_model("base")
|
140 |
+
result = model.transcribe(temp_path, language=source_lang)
|
141 |
+
transcribed_text = result["text"]
|
142 |
+
translated_text = translate_text(transcribed_text, source_lang, target_lang)
|
143 |
+
return {
|
144 |
+
"error": None,
|
145 |
+
"original": transcribed_text,
|
146 |
+
"translated": translated_text,
|
147 |
+
}
|
148 |
+
|
149 |
+
except Exception as e:
|
150 |
+
logger.error(f"خطأ في معالجة الفيديو: {str(e)}")
|
151 |
+
return {
|
152 |
+
"error": f"حدث خطأ: {str(e)}",
|
153 |
+
"original": "",
|
154 |
+
"translated": "",
|
155 |
+
}
|
156 |
+
|
157 |
+
# إنشاء الواجهة
|
158 |
+
def create_ui():
|
159 |
+
with gr.Blocks() as demo:
|
160 |
+
gr.Markdown("# 🎥 تحويل الفيديو إلى نصوص وصوت")
|
161 |
+
with gr.Row():
|
162 |
+
video_input = gr.File(label="📁 رفع فيديو", file_types=["video"])
|
163 |
+
source_lang = gr.Dropdown(
|
164 |
+
choices=list(SUPPORTED_LANGUAGES.keys()),
|
165 |
+
value="en",
|
166 |
+
label="🗣️ لغة الفيديو الأصلية",
|
167 |
+
)
|
168 |
+
target_lang = gr.Dropdown(
|
169 |
+
choices=list(SUPPORTED_LANGUAGES.keys()),
|
170 |
+
value="ar",
|
171 |
+
label="🌐 لغة الترجمة",
|
172 |
+
)
|
173 |
+
process_btn = gr.Button("🎯 معالجة الفيديو")
|
174 |
+
|
175 |
+
with gr.Tabs():
|
176 |
+
with gr.TabItem("النص الأصلي"):
|
177 |
+
original_text = gr.Textbox(label="النص المستخرج", lines=10)
|
178 |
+
original_audio = gr.Audio(label="الصوت")
|
179 |
+
with gr.TabItem("النص المترجم"):
|
180 |
+
translated_text = gr.Textbox(label="النص المترجم", lines=10)
|
181 |
+
translated_audio = gr.Audio(label="الصوت")
|
182 |
+
|
183 |
+
def update_ui(video, src_lang, tgt_lang):
|
184 |
+
result = process_video(video, src_lang, tgt_lang)
|
185 |
+
return {
|
186 |
+
original_text: result["original"],
|
187 |
+
translated_text: result["translated"],
|
188 |
+
}
|
189 |
+
|
190 |
+
process_btn.click(
|
191 |
+
fn=update_ui,
|
192 |
+
inputs=[video_input, source_lang, target_lang],
|
193 |
+
outputs=[original_text, translated_text],
|
194 |
+
)
|
195 |
+
|
196 |
+
return demo
|
197 |
+
|
198 |
+
|
199 |
+
if __name__ == "__main__":
|
200 |
+
demo = create_ui()
|
201 |
+
demo.launch()
|