Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -3,46 +3,6 @@ import tempfile
|
|
| 3 |
import gradio as gr
|
| 4 |
import re
|
| 5 |
import sys
|
| 6 |
-
import shutil
|
| 7 |
-
import importlib.util
|
| 8 |
-
import time
|
| 9 |
-
import random
|
| 10 |
-
|
| 11 |
-
def check_requirements():
|
| 12 |
-
"""Check if all required packages are installed and return status"""
|
| 13 |
-
requirements_status = []
|
| 14 |
-
|
| 15 |
-
packages = [
|
| 16 |
-
('gradio', 'gradio'),
|
| 17 |
-
('yt-dlp', 'yt_dlp'),
|
| 18 |
-
('openai-whisper', 'whisper'),
|
| 19 |
-
('torch', 'torch'),
|
| 20 |
-
('torchaudio', 'torchaudio'),
|
| 21 |
-
('numpy', 'numpy'),
|
| 22 |
-
('regex', 'regex'),
|
| 23 |
-
]
|
| 24 |
-
|
| 25 |
-
for package_name, import_name in packages:
|
| 26 |
-
try:
|
| 27 |
-
spec = importlib.util.find_spec(import_name)
|
| 28 |
-
if spec is None:
|
| 29 |
-
requirements_status.append(f"β {package_name}: Not found")
|
| 30 |
-
continue
|
| 31 |
-
|
| 32 |
-
module = importlib.import_module(import_name)
|
| 33 |
-
version = getattr(module, '__version__', 'Unknown version')
|
| 34 |
-
requirements_status.append(f"β
{package_name}: {version}")
|
| 35 |
-
|
| 36 |
-
except ImportError as e:
|
| 37 |
-
requirements_status.append(f"β {package_name}: Import error - {str(e)}")
|
| 38 |
-
except Exception as e:
|
| 39 |
-
requirements_status.append(f"β οΈ {package_name}: Found but error - {str(e)}")
|
| 40 |
-
|
| 41 |
-
# Add Python info
|
| 42 |
-
requirements_status.append(f"\nπ Python: {sys.version}")
|
| 43 |
-
requirements_status.append(f"π Python executable: {sys.executable}")
|
| 44 |
-
|
| 45 |
-
return "\n".join(requirements_status)
|
| 46 |
|
| 47 |
# Try to import required packages with error handling
|
| 48 |
try:
|
|
@@ -52,317 +12,175 @@ except ImportError as e:
|
|
| 52 |
YT_DLP_AVAILABLE = False
|
| 53 |
print(f"yt-dlp import error: {e}")
|
| 54 |
|
| 55 |
-
# Try multiple whisper import methods
|
| 56 |
-
WHISPER_AVAILABLE = False
|
| 57 |
-
WHISPER_TYPE = None
|
| 58 |
-
|
| 59 |
try:
|
| 60 |
import whisper
|
| 61 |
WHISPER_AVAILABLE = True
|
| 62 |
-
WHISPER_TYPE = "openai-whisper"
|
| 63 |
-
print("Using OpenAI Whisper")
|
| 64 |
except ImportError as e:
|
| 65 |
-
|
| 66 |
-
|
| 67 |
-
from transformers import pipeline
|
| 68 |
-
WHISPER_AVAILABLE = True
|
| 69 |
-
WHISPER_TYPE = "transformers"
|
| 70 |
-
print("Using Transformers Whisper")
|
| 71 |
-
except ImportError as e2:
|
| 72 |
-
print(f"Transformers Whisper import error: {e2}")
|
| 73 |
|
| 74 |
print(f"Python version: {sys.version}")
|
| 75 |
-
print(f"Python executable: {sys.executable}")
|
| 76 |
print(f"yt-dlp available: {YT_DLP_AVAILABLE}")
|
| 77 |
-
print(f"whisper available: {WHISPER_AVAILABLE}
|
| 78 |
|
| 79 |
-
def
|
| 80 |
-
"""Get
|
| 81 |
-
if
|
| 82 |
-
|
| 83 |
-
|
| 84 |
-
|
| 85 |
-
|
| 86 |
-
|
| 87 |
-
|
| 88 |
-
|
| 89 |
-
|
| 90 |
-
|
| 91 |
-
if cookies_file_path and os.path.exists(cookies_file_path):
|
| 92 |
-
ydl_opts['cookiefile'] = cookies_file_path
|
| 93 |
-
|
| 94 |
-
with YoutubeDL(ydl_opts) as ydl:
|
| 95 |
-
try:
|
| 96 |
-
info = ydl.extract_info(url, download=False)
|
| 97 |
-
return {
|
| 98 |
-
'title': info.get('title', 'Unknown'),
|
| 99 |
-
'duration': info.get('duration', 0),
|
| 100 |
-
'availability': info.get('availability', 'unknown'),
|
| 101 |
-
'live_status': info.get('live_status', 'unknown'),
|
| 102 |
-
}
|
| 103 |
-
except Exception as e:
|
| 104 |
-
return {'error': str(e)}
|
| 105 |
|
| 106 |
-
def download_audio(url
|
| 107 |
-
"""Download audio from YouTube URL
|
| 108 |
if not YT_DLP_AVAILABLE:
|
| 109 |
raise Exception("yt-dlp is not available. Please check the installation.")
|
| 110 |
|
| 111 |
try:
|
| 112 |
-
# First, try to get video info
|
| 113 |
-
video_info = get_video_info(url, cookies_file_path)
|
| 114 |
-
if 'error' in video_info:
|
| 115 |
-
raise Exception(f"Video info error: {video_info['error']}")
|
| 116 |
-
|
| 117 |
-
print(f"Video title: {video_info.get('title', 'Unknown')}")
|
| 118 |
-
print(f"Video duration: {video_info.get('duration', 0)} seconds")
|
| 119 |
-
print(f"Video availability: {video_info.get('availability', 'unknown')}")
|
| 120 |
-
|
| 121 |
# Create a temporary directory for downloads
|
| 122 |
temp_dir = tempfile.mkdtemp()
|
| 123 |
output_path = os.path.join(temp_dir, "audio")
|
| 124 |
|
| 125 |
-
#
|
|
|
|
|
|
|
|
|
|
| 126 |
ydl_opts = {
|
| 127 |
-
'format': 'bestaudio[ext=m4a]/bestaudio
|
| 128 |
'outtmpl': output_path + '.%(ext)s',
|
| 129 |
-
'quiet':
|
| 130 |
-
'no_warnings':
|
| 131 |
-
'extractor_retries': 5,
|
| 132 |
-
'fragment_retries': 5,
|
| 133 |
-
'retry_sleep_functions': {'http': lambda n: min(2 ** n, 60)},
|
| 134 |
-
'socket_timeout': 30,
|
| 135 |
-
'http_chunk_size': 10485760, # 10MB chunks
|
| 136 |
-
'writeinfojson': False,
|
| 137 |
-
'writesubtitles': False,
|
| 138 |
-
'writeautomaticsub': False,
|
| 139 |
-
'geo_bypass': True,
|
| 140 |
-
'geo_bypass_country': 'US',
|
| 141 |
'extract_flat': False,
|
| 142 |
'ignoreerrors': False,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 143 |
}
|
| 144 |
|
| 145 |
-
#
|
| 146 |
-
if
|
| 147 |
-
ydl_opts['cookiefile'] =
|
| 148 |
-
print(f"
|
| 149 |
else:
|
| 150 |
-
print("
|
| 151 |
-
|
| 152 |
-
# Always add enhanced headers
|
| 153 |
-
ydl_opts.update({
|
| 154 |
-
'user_agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36',
|
| 155 |
-
'referer': 'https://www.youtube.com/',
|
| 156 |
-
'headers': {
|
| 157 |
-
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36',
|
| 158 |
-
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8',
|
| 159 |
-
'Accept-Language': 'en-US,en;q=0.9',
|
| 160 |
-
'Accept-Encoding': 'gzip, deflate, br',
|
| 161 |
-
'DNT': '1',
|
| 162 |
-
'Connection': 'keep-alive',
|
| 163 |
-
'Upgrade-Insecure-Requests': '1',
|
| 164 |
-
'Sec-Fetch-Dest': 'document',
|
| 165 |
-
'Sec-Fetch-Mode': 'navigate',
|
| 166 |
-
'Sec-Fetch-Site': 'none',
|
| 167 |
-
'Sec-Fetch-User': '?1',
|
| 168 |
-
'sec-ch-ua': '"Not_A Brand";v="8", "Chromium";v="120", "Google Chrome";v="120"',
|
| 169 |
-
'sec-ch-ua-mobile': '?0',
|
| 170 |
-
'sec-ch-ua-platform': '"Windows"',
|
| 171 |
-
}
|
| 172 |
-
})
|
| 173 |
-
|
| 174 |
-
# Add random delay to avoid rate limiting
|
| 175 |
-
time.sleep(random.uniform(1, 3))
|
| 176 |
|
| 177 |
with YoutubeDL(ydl_opts) as ydl:
|
| 178 |
-
|
| 179 |
-
info_dict = ydl.extract_info(url, download=
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 180 |
|
| 181 |
# Find the downloaded file
|
| 182 |
-
for ext in ['.m4a', '.webm', '.mp4', '.mp3'
|
| 183 |
potential_file = output_path + ext
|
| 184 |
if os.path.exists(potential_file):
|
| 185 |
-
print(f"
|
| 186 |
return potential_file
|
| 187 |
|
| 188 |
-
|
| 189 |
-
print(f"Files in temp directory: {os.listdir(temp_dir)}")
|
| 190 |
-
raise FileNotFoundError("Downloaded audio file not found")
|
| 191 |
|
| 192 |
except Exception as e:
|
| 193 |
-
error_msg = str(e)
|
| 194 |
-
|
| 195 |
-
|
| 196 |
-
|
| 197 |
-
raise Exception(
|
| 198 |
-
|
| 199 |
-
|
| 200 |
-
|
| 201 |
-
|
| 202 |
-
β’ Video is geo-blocked in your region
|
| 203 |
-
β’ Video has age restrictions
|
| 204 |
-
β’ Video requires sign-in to view
|
| 205 |
-
β’ Copyright restrictions
|
| 206 |
-
|
| 207 |
-
Solutions to try:
|
| 208 |
-
1. Verify the video URL is correct and accessible
|
| 209 |
-
2. Try a different public video
|
| 210 |
-
3. Check if the video works in your browser
|
| 211 |
-
4. If using a playlist URL, try the direct video URL instead
|
| 212 |
-
5. For age-restricted videos, ensure cookies are from a logged-in account
|
| 213 |
-
|
| 214 |
-
Original error: {str(e)}
|
| 215 |
-
""")
|
| 216 |
-
elif "403" in error_msg or "forbidden" in error_msg:
|
| 217 |
-
raise Exception(f"""
|
| 218 |
-
β Access Forbidden (403): YouTube blocked the request.
|
| 219 |
-
|
| 220 |
-
Solutions:
|
| 221 |
-
1. **Upload fresh cookies.txt file** (most important)
|
| 222 |
-
2. Get cookies from a logged-in YouTube account
|
| 223 |
-
3. Try again after a few minutes (rate limiting)
|
| 224 |
-
4. Use a different network/VPN if possible
|
| 225 |
-
|
| 226 |
-
How to get fresh cookies:
|
| 227 |
-
β’ Visit YouTube while logged in
|
| 228 |
-
β’ Use browser extension to export cookies
|
| 229 |
-
β’ Upload the newest cookies.txt file
|
| 230 |
-
|
| 231 |
-
Original error: {str(e)}
|
| 232 |
-
""")
|
| 233 |
-
elif "429" in error_msg or "rate limit" in error_msg:
|
| 234 |
-
raise Exception(f"""
|
| 235 |
-
β Rate Limited (429): Too many requests.
|
| 236 |
-
|
| 237 |
-
Solutions:
|
| 238 |
-
1. Wait 10-15 minutes before trying again
|
| 239 |
-
2. Upload fresh cookies.txt file
|
| 240 |
-
3. Try a different video
|
| 241 |
-
4. Use a different network if possible
|
| 242 |
-
|
| 243 |
-
Original error: {str(e)}
|
| 244 |
-
""")
|
| 245 |
else:
|
| 246 |
-
raise Exception(f"Failed to download audio: {
|
| 247 |
|
| 248 |
def transcribe_audio(file_path):
|
| 249 |
"""Transcribe audio file using Whisper"""
|
| 250 |
if not WHISPER_AVAILABLE:
|
| 251 |
-
raise Exception("OpenAI Whisper is not available. Please
|
| 252 |
|
| 253 |
try:
|
| 254 |
-
|
| 255 |
-
|
| 256 |
-
|
| 257 |
-
|
| 258 |
-
file_path,
|
| 259 |
-
language="en", # Specify English for better performance
|
| 260 |
-
task="transcribe",
|
| 261 |
-
verbose=False,
|
| 262 |
-
fp16=False, # Better compatibility
|
| 263 |
-
temperature=0.0, # More deterministic
|
| 264 |
-
)
|
| 265 |
-
return result["text"]
|
| 266 |
-
|
| 267 |
-
elif WHISPER_TYPE == "transformers":
|
| 268 |
-
# Use Transformers Whisper
|
| 269 |
-
from transformers import pipeline
|
| 270 |
-
transcriber = pipeline(
|
| 271 |
-
"automatic-speech-recognition",
|
| 272 |
-
model="openai/whisper-base",
|
| 273 |
-
device=-1 # Use CPU for better compatibility
|
| 274 |
-
)
|
| 275 |
-
result = transcriber(file_path, return_timestamps=False)
|
| 276 |
-
return result["text"]
|
| 277 |
-
|
| 278 |
-
else:
|
| 279 |
-
raise Exception("No compatible Whisper installation found")
|
| 280 |
-
|
| 281 |
except Exception as e:
|
| 282 |
raise Exception(f"Failed to transcribe audio: {str(e)}")
|
| 283 |
|
| 284 |
-
def
|
| 285 |
-
"""
|
| 286 |
try:
|
| 287 |
stock_info = []
|
| 288 |
|
| 289 |
-
#
|
| 290 |
-
stock_patterns =
|
| 291 |
-
|
| 292 |
-
|
| 293 |
-
|
| 294 |
-
|
| 295 |
-
|
| 296 |
-
'market_terms': r'\b(?:earnings|revenue|profit|loss|growth|dividend|yield|PE ratio|market cap|volume)\b',
|
| 297 |
-
}
|
| 298 |
-
|
| 299 |
-
# Extract information
|
| 300 |
-
symbols = re.findall(stock_patterns['symbols'], text, re.IGNORECASE)
|
| 301 |
-
prices = re.findall(stock_patterns['prices'], text)
|
| 302 |
-
percentages = re.findall(stock_patterns['percentages'], text)
|
| 303 |
-
actions = re.findall(stock_patterns['actions'], text, re.IGNORECASE)
|
| 304 |
-
companies = re.findall(stock_patterns['companies'], text)
|
| 305 |
-
market_terms = re.findall(stock_patterns['market_terms'], text, re.IGNORECASE)
|
| 306 |
|
| 307 |
-
#
|
| 308 |
-
|
|
|
|
|
|
|
|
|
|
| 309 |
|
| 310 |
-
|
| 311 |
-
|
| 312 |
|
| 313 |
if companies:
|
| 314 |
-
|
| 315 |
-
if filtered_companies:
|
| 316 |
-
result += f"π’ **Companies Mentioned**: {', '.join(filtered_companies[:10])}\n\n"
|
| 317 |
|
| 318 |
-
if
|
| 319 |
-
result += f"
|
| 320 |
|
| 321 |
-
if
|
| 322 |
-
result += f"
|
| 323 |
|
| 324 |
if actions:
|
| 325 |
-
result += f"
|
| 326 |
|
| 327 |
-
|
| 328 |
-
result += f"π **Market Terms**: {', '.join(set(market_terms[:10]))}\n\n"
|
| 329 |
-
|
| 330 |
-
# Look for recommendation sentences
|
| 331 |
-
sentences = [s.strip() for s in text.split('.') if s.strip()]
|
| 332 |
recommendations = []
|
| 333 |
-
|
| 334 |
for sentence in sentences:
|
| 335 |
-
|
| 336 |
-
|
| 337 |
-
|
| 338 |
-
recommendations.append(sentence)
|
| 339 |
|
| 340 |
if recommendations:
|
| 341 |
-
result += "π―
|
| 342 |
-
for
|
| 343 |
-
result += f"
|
| 344 |
-
result += "\n"
|
| 345 |
|
| 346 |
-
|
| 347 |
-
|
| 348 |
-
result += "
|
| 349 |
-
result += "
|
| 350 |
-
result += "β’
|
| 351 |
-
result += "β’
|
| 352 |
-
result += "β’ Content is not in English\n"
|
| 353 |
-
result += "β’ General market discussion without specific recommendations\n"
|
| 354 |
-
else:
|
| 355 |
-
result += "β
**Analysis Complete** - Please verify all information independently!\n"
|
| 356 |
-
|
| 357 |
-
result += "\n" + "="*50 + "\n"
|
| 358 |
-
result += "β οΈ **DISCLAIMER**: This is automated extraction for educational purposes only.\n"
|
| 359 |
-
result += "Always conduct your own research before making investment decisions!\n"
|
| 360 |
-
result += "="*50
|
| 361 |
|
| 362 |
return result
|
| 363 |
|
| 364 |
except Exception as e:
|
| 365 |
-
return f"
|
| 366 |
|
| 367 |
def cleanup_file(file_path):
|
| 368 |
"""Clean up temporary files"""
|
|
@@ -377,487 +195,188 @@ def cleanup_file(file_path):
|
|
| 377 |
except:
|
| 378 |
pass
|
| 379 |
|
| 380 |
-
def
|
| 381 |
-
"""
|
| 382 |
-
|
| 383 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 384 |
|
|
|
|
| 385 |
try:
|
| 386 |
-
|
| 387 |
-
|
| 388 |
-
|
| 389 |
-
|
| 390 |
-
|
| 391 |
-
|
| 392 |
-
# Validate cookies file
|
| 393 |
-
with open(temp_cookies_path, 'r', encoding='utf-8') as f:
|
| 394 |
-
content = f.read()
|
| 395 |
-
if 'youtube.com' not in content.lower():
|
| 396 |
-
print("β οΈ Warning: cookies file might not contain YouTube cookies")
|
| 397 |
-
|
| 398 |
-
print(f"β
Cookies file processed: {temp_cookies_path}")
|
| 399 |
-
return temp_cookies_path
|
| 400 |
except Exception as e:
|
| 401 |
-
|
| 402 |
-
return None
|
| 403 |
-
|
| 404 |
-
def validate_youtube_url(url):
|
| 405 |
-
"""Validate YouTube URL format"""
|
| 406 |
-
if not url or not url.strip():
|
| 407 |
-
return False, "Please provide a YouTube URL"
|
| 408 |
-
|
| 409 |
-
url = url.strip()
|
| 410 |
-
youtube_patterns = [
|
| 411 |
-
r'(?:https?://)?(?:www\.)?youtube\.com/watch\?v=[\w-]+',
|
| 412 |
-
r'(?:https?://)?(?:www\.)?youtu\.be/[\w-]+',
|
| 413 |
-
r'(?:https?://)?(?:www\.)?youtube\.com/embed/[\w-]+',
|
| 414 |
-
r'(?:https?://)?(?:m\.)?youtube\.com/watch\?v=[\w-]+',
|
| 415 |
-
]
|
| 416 |
|
| 417 |
-
|
| 418 |
-
|
| 419 |
-
|
|
|
|
|
|
|
|
|
|
| 420 |
|
| 421 |
-
return
|
| 422 |
|
| 423 |
-
def process_video(url,
|
| 424 |
-
"""Main function to process YouTube video
|
| 425 |
-
|
| 426 |
-
# Detailed debugging info
|
| 427 |
-
debug_info = []
|
| 428 |
-
debug_info.append(f"π Starting process at {time.strftime('%H:%M:%S')}")
|
| 429 |
-
debug_info.append(f"π‘ Python version: {sys.version.split()[0]}")
|
| 430 |
-
debug_info.append(f"π¦ yt-dlp available: {YT_DLP_AVAILABLE}")
|
| 431 |
-
debug_info.append(f"ποΈ Whisper available: {WHISPER_AVAILABLE} (type: {WHISPER_TYPE})")
|
| 432 |
|
| 433 |
# Check if required packages are available
|
| 434 |
if not YT_DLP_AVAILABLE:
|
| 435 |
-
|
| 436 |
-
error_msg += "SOLUTION: Install yt-dlp using:\n"
|
| 437 |
-
error_msg += "pip install yt-dlp\n\n"
|
| 438 |
-
error_msg += "DEBUG INFO:\n" + "\n".join(debug_info)
|
| 439 |
-
return error_msg, "", "β Missing yt-dlp"
|
| 440 |
|
| 441 |
if not WHISPER_AVAILABLE:
|
| 442 |
-
|
| 443 |
-
error_msg += "SOLUTION: Install Whisper using:\n"
|
| 444 |
-
error_msg += "pip install openai-whisper\n"
|
| 445 |
-
error_msg += "OR\n"
|
| 446 |
-
error_msg += "pip install transformers torch torchaudio\n\n"
|
| 447 |
-
error_msg += "DEBUG INFO:\n" + "\n".join(debug_info)
|
| 448 |
-
return error_msg, "", "β Missing Whisper"
|
| 449 |
|
| 450 |
-
|
| 451 |
-
|
| 452 |
-
if not is_valid:
|
| 453 |
-
error_msg = f"β ERROR: {validation_msg}\n\n"
|
| 454 |
-
error_msg += f"PROVIDED URL: {url}\n\n"
|
| 455 |
-
error_msg += "VALID URL FORMATS:\n"
|
| 456 |
-
error_msg += "οΏ½οΏ½ https://www.youtube.com/watch?v=VIDEO_ID\n"
|
| 457 |
-
error_msg += "β’ https://youtu.be/VIDEO_ID\n"
|
| 458 |
-
error_msg += "β’ https://www.youtube.com/embed/VIDEO_ID\n\n"
|
| 459 |
-
error_msg += "DEBUG INFO:\n" + "\n".join(debug_info)
|
| 460 |
-
return error_msg, "", "β Invalid URL"
|
| 461 |
|
| 462 |
audio_path = None
|
| 463 |
-
cookies_temp_path = None
|
| 464 |
-
|
| 465 |
try:
|
| 466 |
-
|
| 467 |
-
|
| 468 |
-
|
| 469 |
-
# Process cookies file if provided
|
| 470 |
-
progress(0.1, desc="πͺ Processing cookies...")
|
| 471 |
-
cookies_temp_path = process_cookies_file(cookies_file)
|
| 472 |
-
|
| 473 |
-
if cookies_temp_path:
|
| 474 |
-
debug_info.append(f"β
Cookies processed: {cookies_temp_path}")
|
| 475 |
-
else:
|
| 476 |
-
debug_info.append("β οΈ No cookies provided - this may cause access errors")
|
| 477 |
-
|
| 478 |
-
status_msg = "β
Cookies loaded" if cookies_temp_path else "β οΈ No cookies (may encounter restrictions)"
|
| 479 |
-
|
| 480 |
-
# First, try to get video info for debugging
|
| 481 |
-
progress(0.15, desc="π Checking video accessibility...")
|
| 482 |
-
try:
|
| 483 |
-
video_info = get_video_info(url, cookies_temp_path)
|
| 484 |
-
if 'error' in video_info:
|
| 485 |
-
debug_info.append(f"β Video info error: {video_info['error']}")
|
| 486 |
-
raise Exception(f"Video accessibility check failed: {video_info['error']}")
|
| 487 |
-
else:
|
| 488 |
-
debug_info.append(f"β
Video info: {video_info}")
|
| 489 |
-
except Exception as e:
|
| 490 |
-
debug_info.append(f"β Video info check failed: {str(e)}")
|
| 491 |
-
# Continue anyway, but log the issue
|
| 492 |
|
| 493 |
# Download audio
|
| 494 |
-
progress(0.
|
| 495 |
-
|
| 496 |
-
audio_path = download_audio(url, cookies_temp_path)
|
| 497 |
-
debug_info.append(f"β
Audio downloaded: {audio_path}")
|
| 498 |
-
|
| 499 |
-
# Check if audio file exists and get size
|
| 500 |
-
if audio_path and os.path.exists(audio_path):
|
| 501 |
-
file_size = os.path.getsize(audio_path)
|
| 502 |
-
debug_info.append(f"π Audio file size: {file_size/1024/1024:.2f} MB")
|
| 503 |
-
else:
|
| 504 |
-
raise Exception("Audio file not found after download")
|
| 505 |
|
| 506 |
# Transcribe audio
|
| 507 |
-
progress(0.
|
| 508 |
-
debug_info.append("π Starting transcription...")
|
| 509 |
transcript = transcribe_audio(audio_path)
|
| 510 |
-
debug_info.append(f"β
Transcription completed: {len(transcript)} characters")
|
| 511 |
|
| 512 |
if not transcript.strip():
|
| 513 |
-
|
| 514 |
-
error_msg += "POSSIBLE CAUSES:\n"
|
| 515 |
-
error_msg += "β’ Video has no audio track\n"
|
| 516 |
-
error_msg += "β’ Audio is too quiet or unclear\n"
|
| 517 |
-
error_msg += "β’ Video is not in English\n"
|
| 518 |
-
error_msg += "β’ Audio file is corrupted\n\n"
|
| 519 |
-
error_msg += "DEBUG INFO:\n" + "\n".join(debug_info)
|
| 520 |
-
return error_msg, "", "β No speech detected"
|
| 521 |
|
| 522 |
# Extract stock information
|
| 523 |
-
progress(0.
|
| 524 |
-
|
| 525 |
-
stock_details = extract_stock_info_enhanced(transcript)
|
| 526 |
-
debug_info.append("β
Stock analysis completed")
|
| 527 |
-
|
| 528 |
-
progress(1.0, desc="β
Complete!")
|
| 529 |
|
| 530 |
-
|
| 531 |
-
|
| 532 |
-
debug_section += "π DEBUG INFORMATION\n"
|
| 533 |
-
debug_section += "="*50 + "\n"
|
| 534 |
-
debug_section += "\n".join(debug_info)
|
| 535 |
-
|
| 536 |
-
return transcript + debug_section, stock_details, "β
Processing completed successfully"
|
| 537 |
|
| 538 |
except Exception as e:
|
| 539 |
-
error_msg = f"
|
| 540 |
-
error_msg
|
| 541 |
-
error_msg += f"ERROR TYPE: {type(e).__name__}\n\n"
|
| 542 |
-
|
| 543 |
-
# Add context based on where the error occurred
|
| 544 |
-
if "download" in str(e).lower():
|
| 545 |
-
error_msg += "π§ DOWNLOAD TROUBLESHOOTING:\n"
|
| 546 |
-
error_msg += "β’ Check if video URL is accessible in browser\n"
|
| 547 |
-
error_msg += "β’ Upload fresh cookies.txt file\n"
|
| 548 |
-
error_msg += "β’ Try a different video\n"
|
| 549 |
-
error_msg += "β’ Wait 10-15 minutes if rate limited\n\n"
|
| 550 |
-
elif "transcribe" in str(e).lower():
|
| 551 |
-
error_msg += "π§ TRANSCRIPTION TROUBLESHOOTING:\n"
|
| 552 |
-
error_msg += "β’ Check if audio file was downloaded properly\n"
|
| 553 |
-
error_msg += "β’ Ensure video has clear audio\n"
|
| 554 |
-
error_msg += "β’ Try a shorter video\n\n"
|
| 555 |
-
|
| 556 |
-
error_msg += "π PROCESSING STEPS COMPLETED:\n"
|
| 557 |
-
error_msg += "\n".join(debug_info)
|
| 558 |
-
|
| 559 |
-
return error_msg, "", f"β Error: {type(e).__name__}"
|
| 560 |
|
| 561 |
finally:
|
| 562 |
# Clean up temporary files
|
| 563 |
-
|
| 564 |
-
debug_info.append(f"ποΈ Cleaning up: {audio_path}")
|
| 565 |
-
cleanup_file(audio_path)
|
| 566 |
-
if cookies_temp_path:
|
| 567 |
-
debug_info.append(f"ποΈ Cleaning up: {cookies_temp_path}")
|
| 568 |
-
cleanup_file(cookies_temp_path)
|
| 569 |
|
| 570 |
-
# Create Gradio interface
|
| 571 |
with gr.Blocks(
|
| 572 |
-
title="
|
| 573 |
theme=gr.themes.Soft(),
|
| 574 |
css="""
|
| 575 |
.gradio-container {
|
| 576 |
max-width: 1200px;
|
| 577 |
margin: auto;
|
| 578 |
-
font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
|
| 579 |
-
}
|
| 580 |
-
.status-box {
|
| 581 |
-
padding: 12px;
|
| 582 |
-
border-radius: 8px;
|
| 583 |
-
margin: 10px 0;
|
| 584 |
-
border: 1px solid #ddd;
|
| 585 |
-
}
|
| 586 |
-
.warning-box {
|
| 587 |
-
background-color: #fff3cd;
|
| 588 |
-
border-color: #ffeaa7;
|
| 589 |
-
color: #856404;
|
| 590 |
-
}
|
| 591 |
-
.success-box {
|
| 592 |
-
background-color: #d4edda;
|
| 593 |
-
border-color: #c3e6cb;
|
| 594 |
-
color: #155724;
|
| 595 |
-
}
|
| 596 |
-
.error-box {
|
| 597 |
-
background-color: #f8d7da;
|
| 598 |
-
border-color: #f5c6cb;
|
| 599 |
-
color: #721c24;
|
| 600 |
}
|
| 601 |
"""
|
| 602 |
) as demo:
|
| 603 |
|
| 604 |
gr.Markdown("""
|
| 605 |
-
# π
|
| 606 |
|
| 607 |
-
|
| 608 |
|
| 609 |
-
|
| 610 |
-
1.
|
| 611 |
-
2.
|
| 612 |
-
3.
|
| 613 |
-
4. **Extracts** stock symbols, prices, and recommendations
|
| 614 |
|
| 615 |
-
|
| 616 |
""")
|
| 617 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 618 |
with gr.Row():
|
| 619 |
with gr.Column(scale=1):
|
| 620 |
-
|
| 621 |
-
|
| 622 |
-
|
| 623 |
-
|
| 624 |
-
|
| 625 |
-
variant="secondary",
|
| 626 |
-
size="sm"
|
| 627 |
-
)
|
| 628 |
-
|
| 629 |
-
requirements_output = gr.Textbox(
|
| 630 |
-
label="π System Requirements Status",
|
| 631 |
-
lines=8,
|
| 632 |
-
max_lines=15,
|
| 633 |
-
interactive=False,
|
| 634 |
-
visible=False
|
| 635 |
-
)
|
| 636 |
|
| 637 |
-
|
| 638 |
-
|
| 639 |
-
|
| 640 |
-
|
| 641 |
-
|
| 642 |
-
|
| 643 |
-
|
| 644 |
-
|
| 645 |
-
|
| 646 |
-
|
| 647 |
-
|
| 648 |
-
|
| 649 |
-
|
| 650 |
-
lines=5,
|
| 651 |
-
visible=False,
|
| 652 |
-
interactive=False
|
| 653 |
-
)
|
| 654 |
-
|
| 655 |
-
# Cookies upload with better instructions
|
| 656 |
-
cookies_input = gr.File(
|
| 657 |
-
label="πͺ Upload Cookies File (cookies.txt) - HIGHLY RECOMMENDED",
|
| 658 |
-
file_types=[".txt"],
|
| 659 |
-
file_count="single"
|
| 660 |
-
)
|
| 661 |
-
|
| 662 |
-
with gr.Accordion("π How to Get Cookies (Click to expand)", open=False):
|
| 663 |
-
gr.Markdown("""
|
| 664 |
-
**Why cookies are needed:** YouTube blocks most automated requests without proper authentication.
|
| 665 |
-
|
| 666 |
-
**Step-by-step instructions:**
|
| 667 |
-
1. **Install browser extension:**
|
| 668 |
-
- Chrome: "Get cookies.txt LOCALLY" or "cookies.txt"
|
| 669 |
-
- Firefox: "cookies.txt" or "Export Cookies"
|
| 670 |
-
|
| 671 |
-
2. **Get cookies:**
|
| 672 |
-
- Visit YouTube.com (log in if needed)
|
| 673 |
-
- Click the extension icon
|
| 674 |
-
- Select "Export for youtube.com"
|
| 675 |
-
- Download the cookies.txt file
|
| 676 |
-
|
| 677 |
-
3. **Upload here:** Use the file upload above
|
| 678 |
-
|
| 679 |
-
**β οΈ Without cookies, you'll get "403 Forbidden" or "Video unavailable" errors**
|
| 680 |
-
""")
|
| 681 |
-
|
| 682 |
-
url_input = gr.Textbox(
|
| 683 |
-
label="πΊ YouTube Video URL",
|
| 684 |
-
placeholder="https://www.youtube.com/watch?v=VIDEO_ID",
|
| 685 |
-
lines=2,
|
| 686 |
-
info="Paste the full YouTube video URL here"
|
| 687 |
-
)
|
| 688 |
-
|
| 689 |
-
process_btn = gr.Button(
|
| 690 |
-
"π Extract Stock Information",
|
| 691 |
-
variant="primary",
|
| 692 |
-
size="lg"
|
| 693 |
-
)
|
| 694 |
-
|
| 695 |
-
# Status display
|
| 696 |
-
status_output = gr.Textbox(
|
| 697 |
-
label="π Status",
|
| 698 |
-
lines=3,
|
| 699 |
-
interactive=False,
|
| 700 |
-
info="Current processing status"
|
| 701 |
-
)
|
| 702 |
|
| 703 |
-
# Output section
|
| 704 |
with gr.Row():
|
| 705 |
with gr.Column():
|
| 706 |
transcript_output = gr.Textbox(
|
| 707 |
label="π Full Transcript",
|
| 708 |
-
lines=
|
| 709 |
-
max_lines=
|
| 710 |
-
show_copy_button=True
|
| 711 |
-
info="Complete transcription of the video audio"
|
| 712 |
)
|
| 713 |
|
| 714 |
with gr.Column():
|
| 715 |
stock_info_output = gr.Textbox(
|
| 716 |
label="π Extracted Stock Information",
|
| 717 |
-
lines=
|
| 718 |
-
max_lines=
|
| 719 |
-
show_copy_button=True
|
| 720 |
-
info="Parsed stock symbols, prices, and recommendations"
|
| 721 |
)
|
| 722 |
|
| 723 |
-
# Example and troubleshooting section
|
| 724 |
-
with gr.Row():
|
| 725 |
-
with gr.Column():
|
| 726 |
-
gr.Markdown("### π Example URLs")
|
| 727 |
-
gr.Examples(
|
| 728 |
-
examples=[
|
| 729 |
-
["https://www.youtube.com/watch?v=dQw4w9WgXcQ"],
|
| 730 |
-
["https://youtu.be/dQw4w9WgXcQ"],
|
| 731 |
-
],
|
| 732 |
-
inputs=[url_input],
|
| 733 |
-
label="Click to try example URLs (replace with actual financial videos)"
|
| 734 |
-
)
|
| 735 |
-
|
| 736 |
-
# Troubleshooting section
|
| 737 |
-
with gr.Accordion("π§ Troubleshooting Guide", open=False):
|
| 738 |
-
gr.Markdown("""
|
| 739 |
-
### Common Issues and Solutions:
|
| 740 |
-
|
| 741 |
-
**β "Video unavailable" or "Content isn't available":**
|
| 742 |
-
- Video might be private, deleted, or geo-blocked
|
| 743 |
-
- Try a different public financial video
|
| 744 |
-
- Verify the URL works in your browser
|
| 745 |
-
- Check if video requires age verification
|
| 746 |
-
|
| 747 |
-
**β "403 Forbidden" error:**
|
| 748 |
-
- **Upload fresh cookies.txt file** (most common fix)
|
| 749 |
-
- Make sure cookies are from a logged-in YouTube account
|
| 750 |
-
- Try waiting 10-15 minutes (rate limiting)
|
| 751 |
-
|
| 752 |
-
**β "No speech detected":**
|
| 753 |
-
- Video might not have clear audio
|
| 754 |
-
- Try videos with clear narration
|
| 755 |
-
- Check if video is in English
|
| 756 |
-
|
| 757 |
-
**β "No stock information found":**
|
| 758 |
-
- Video might not contain financial content
|
| 759 |
-
- Try videos from financial YouTube channels
|
| 760 |
-
- Look for videos with stock analysis or recommendations
|
| 761 |
-
|
| 762 |
-
### Installation Commands:
|
| 763 |
-
```bash
|
| 764 |
-
# Install all requirements
|
| 765 |
-
pip install gradio yt-dlp openai-whisper torch torchaudio
|
| 766 |
-
|
| 767 |
-
# Alternative whisper installation
|
| 768 |
-
pip install transformers torch torchaudio
|
| 769 |
-
```
|
| 770 |
-
|
| 771 |
-
### Best Practices:
|
| 772 |
-
- Use videos from reputable financial channels
|
| 773 |
-
- Prefer videos under 20 minutes for faster processing
|
| 774 |
-
- Ensure clear audio quality
|
| 775 |
-
- Always verify extracted information independently
|
| 776 |
-
""")
|
| 777 |
-
|
| 778 |
# Event handlers
|
| 779 |
-
def show_requirements():
|
| 780 |
-
status = check_requirements()
|
| 781 |
-
return gr.update(value=status, visible=True)
|
| 782 |
-
|
| 783 |
-
def test_system():
|
| 784 |
-
"""Test system components and return detailed status"""
|
| 785 |
-
test_results = []
|
| 786 |
-
test_results.append("π§ͺ SYSTEM TEST RESULTS")
|
| 787 |
-
test_results.append("="*30)
|
| 788 |
-
|
| 789 |
-
# Test imports
|
| 790 |
-
test_results.append(f"β
yt-dlp: {'Available' if YT_DLP_AVAILABLE else 'NOT AVAILABLE'}")
|
| 791 |
-
test_results.append(f"β
Whisper: {'Available' if WHISPER_AVAILABLE else 'NOT AVAILABLE'} (Type: {WHISPER_TYPE})")
|
| 792 |
-
|
| 793 |
-
# Test yt-dlp functionality
|
| 794 |
-
if YT_DLP_AVAILABLE:
|
| 795 |
-
try:
|
| 796 |
-
from yt_dlp import YoutubeDL
|
| 797 |
-
test_ydl = YoutubeDL({'quiet': True})
|
| 798 |
-
test_results.append("β
yt-dlp: Can create YoutubeDL instance")
|
| 799 |
-
except Exception as e:
|
| 800 |
-
test_results.append(f"β yt-dlp: Error creating instance - {str(e)}")
|
| 801 |
-
|
| 802 |
-
# Test Whisper functionality
|
| 803 |
-
if WHISPER_AVAILABLE:
|
| 804 |
-
try:
|
| 805 |
-
if WHISPER_TYPE == "openai-whisper":
|
| 806 |
-
import whisper
|
| 807 |
-
test_results.append("β
Whisper: OpenAI Whisper can be imported")
|
| 808 |
-
elif WHISPER_TYPE == "transformers":
|
| 809 |
-
from transformers import pipeline
|
| 810 |
-
test_results.append("β
Whisper: Transformers Whisper can be imported")
|
| 811 |
-
except Exception as e:
|
| 812 |
-
test_results.append(f"β Whisper: Error testing - {str(e)}")
|
| 813 |
-
|
| 814 |
-
# Test file operations
|
| 815 |
-
try:
|
| 816 |
-
temp_file = tempfile.mktemp()
|
| 817 |
-
with open(temp_file, 'w') as f:
|
| 818 |
-
f.write("test")
|
| 819 |
-
os.remove(temp_file)
|
| 820 |
-
test_results.append("β
File operations: Working")
|
| 821 |
-
except Exception as e:
|
| 822 |
-
test_results.append(f"β File operations: Error - {str(e)}")
|
| 823 |
-
|
| 824 |
-
test_results.append("\nπ‘ If you see any β errors above, install missing packages:")
|
| 825 |
-
test_results.append("pip install yt-dlp openai-whisper torch torchaudio")
|
| 826 |
-
|
| 827 |
-
return gr.update(value="\n".join(test_results), visible=True)
|
| 828 |
-
|
| 829 |
-
check_req_btn.click(
|
| 830 |
-
fn=show_requirements,
|
| 831 |
-
outputs=[requirements_output]
|
| 832 |
-
)
|
| 833 |
-
|
| 834 |
-
test_btn.click(
|
| 835 |
-
fn=test_system,
|
| 836 |
-
outputs=[test_output]
|
| 837 |
-
)
|
| 838 |
-
|
| 839 |
process_btn.click(
|
| 840 |
fn=process_video,
|
| 841 |
-
inputs=[url_input
|
| 842 |
-
outputs=[transcript_output, stock_info_output
|
| 843 |
show_progress=True
|
| 844 |
)
|
| 845 |
|
| 846 |
-
#
|
| 847 |
-
gr.Markdown(""
|
| 848 |
-
|
| 849 |
-
|
| 850 |
-
|
| 851 |
-
|
| 852 |
-
|
|
|
|
|
|
|
| 853 |
|
| 854 |
-
# Launch configuration for Gradio Cloud
|
| 855 |
if __name__ == "__main__":
|
| 856 |
-
demo.launch(
|
| 857 |
-
server_name="0.0.0.0",
|
| 858 |
-
server_port=7860,
|
| 859 |
-
share=False,
|
| 860 |
-
debug=False,
|
| 861 |
-
show_error=True,
|
| 862 |
-
quiet=False
|
| 863 |
-
)
|
|
|
|
| 3 |
import gradio as gr
|
| 4 |
import re
|
| 5 |
import sys
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
|
| 7 |
# Try to import required packages with error handling
|
| 8 |
try:
|
|
|
|
| 12 |
YT_DLP_AVAILABLE = False
|
| 13 |
print(f"yt-dlp import error: {e}")
|
| 14 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
try:
|
| 16 |
import whisper
|
| 17 |
WHISPER_AVAILABLE = True
|
|
|
|
|
|
|
| 18 |
except ImportError as e:
|
| 19 |
+
WHISPER_AVAILABLE = False
|
| 20 |
+
print(f"whisper import error: {e}")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 21 |
|
| 22 |
print(f"Python version: {sys.version}")
|
|
|
|
| 23 |
print(f"yt-dlp available: {YT_DLP_AVAILABLE}")
|
| 24 |
+
print(f"whisper available: {WHISPER_AVAILABLE}")
|
| 25 |
|
| 26 |
+
def get_cookies_path():
|
| 27 |
+
"""Get the path to cookies.txt file"""
|
| 28 |
+
# Check if cookies.txt exists in the current directory
|
| 29 |
+
if os.path.exists('cookies.txt'):
|
| 30 |
+
return 'cookies.txt'
|
| 31 |
+
# Check in the same directory as the script
|
| 32 |
+
script_dir = os.path.dirname(os.path.abspath(__file__))
|
| 33 |
+
cookies_path = os.path.join(script_dir, 'cookies.txt')
|
| 34 |
+
if os.path.exists(cookies_path):
|
| 35 |
+
return cookies_path
|
| 36 |
+
return None
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 37 |
|
| 38 |
+
def download_audio(url):
|
| 39 |
+
"""Download audio from YouTube URL and return the file path"""
|
| 40 |
if not YT_DLP_AVAILABLE:
|
| 41 |
raise Exception("yt-dlp is not available. Please check the installation.")
|
| 42 |
|
| 43 |
try:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 44 |
# Create a temporary directory for downloads
|
| 45 |
temp_dir = tempfile.mkdtemp()
|
| 46 |
output_path = os.path.join(temp_dir, "audio")
|
| 47 |
|
| 48 |
+
# Get cookies path
|
| 49 |
+
cookies_path = get_cookies_path()
|
| 50 |
+
|
| 51 |
+
# Base yt-dlp options
|
| 52 |
ydl_opts = {
|
| 53 |
+
'format': 'bestaudio[ext=m4a]/bestaudio/best',
|
| 54 |
'outtmpl': output_path + '.%(ext)s',
|
| 55 |
+
'quiet': True,
|
| 56 |
+
'no_warnings': True,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 57 |
'extract_flat': False,
|
| 58 |
'ignoreerrors': False,
|
| 59 |
+
# Add user agent to avoid bot detection
|
| 60 |
+
'http_headers': {
|
| 61 |
+
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36'
|
| 62 |
+
},
|
| 63 |
+
# Add additional options to avoid bot detection
|
| 64 |
+
'extractor_retries': 3,
|
| 65 |
+
'fragment_retries': 3,
|
| 66 |
+
'retry_sleep_functions': {'http': lambda n: 2 ** n},
|
| 67 |
}
|
| 68 |
|
| 69 |
+
# Add cookies if available
|
| 70 |
+
if cookies_path:
|
| 71 |
+
ydl_opts['cookiefile'] = cookies_path
|
| 72 |
+
print(f"Using cookies from: {cookies_path}")
|
| 73 |
else:
|
| 74 |
+
print("No cookies.txt found - proceeding without cookies")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 75 |
|
| 76 |
with YoutubeDL(ydl_opts) as ydl:
|
| 77 |
+
# Extract info first to check if video is available
|
| 78 |
+
info_dict = ydl.extract_info(url, download=False)
|
| 79 |
+
|
| 80 |
+
# Check if video is available
|
| 81 |
+
if info_dict.get('availability') == 'private':
|
| 82 |
+
raise Exception("Video is private")
|
| 83 |
+
elif info_dict.get('availability') == 'premium_only':
|
| 84 |
+
raise Exception("Video requires premium subscription")
|
| 85 |
+
elif info_dict.get('live_status') == 'is_live':
|
| 86 |
+
raise Exception("Cannot download live streams")
|
| 87 |
+
|
| 88 |
+
# Download the audio
|
| 89 |
+
ydl.download([url])
|
| 90 |
|
| 91 |
# Find the downloaded file
|
| 92 |
+
for ext in ['.m4a', '.webm', '.mp4', '.mp3']:
|
| 93 |
potential_file = output_path + ext
|
| 94 |
if os.path.exists(potential_file):
|
| 95 |
+
print(f"Successfully downloaded: {potential_file}")
|
| 96 |
return potential_file
|
| 97 |
|
| 98 |
+
raise FileNotFoundError(f"Downloaded audio file not found")
|
|
|
|
|
|
|
| 99 |
|
| 100 |
except Exception as e:
|
| 101 |
+
error_msg = str(e)
|
| 102 |
+
if "Sign in to confirm your age" in error_msg:
|
| 103 |
+
raise Exception("Video is age-restricted. Please use a different video or update your cookies.")
|
| 104 |
+
elif "Private video" in error_msg:
|
| 105 |
+
raise Exception("Video is private and cannot be accessed.")
|
| 106 |
+
elif "This video is unavailable" in error_msg:
|
| 107 |
+
raise Exception("Video is unavailable or has been removed.")
|
| 108 |
+
elif "blocked" in error_msg.lower():
|
| 109 |
+
raise Exception("Access to this video is blocked. Try using updated cookies or a different video.")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 110 |
else:
|
| 111 |
+
raise Exception(f"Failed to download audio: {error_msg}")
|
| 112 |
|
| 113 |
def transcribe_audio(file_path):
|
| 114 |
"""Transcribe audio file using Whisper"""
|
| 115 |
if not WHISPER_AVAILABLE:
|
| 116 |
+
raise Exception("OpenAI Whisper is not available. Please check the installation.")
|
| 117 |
|
| 118 |
try:
|
| 119 |
+
# Use the smallest model to reduce memory usage
|
| 120 |
+
model = whisper.load_model("tiny")
|
| 121 |
+
result = model.transcribe(file_path)
|
| 122 |
+
return result["text"]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 123 |
except Exception as e:
|
| 124 |
raise Exception(f"Failed to transcribe audio: {str(e)}")
|
| 125 |
|
| 126 |
+
def extract_stock_info_simple(text):
|
| 127 |
+
"""Extract stock information using simple pattern matching"""
|
| 128 |
try:
|
| 129 |
stock_info = []
|
| 130 |
|
| 131 |
+
# Simple patterns to look for stock-related information
|
| 132 |
+
stock_patterns = [
|
| 133 |
+
r'\b[A-Z]{1,5}\b(?:\s+stock|\s+shares|\s+symbol)', # Stock symbols
|
| 134 |
+
r'(?:buy|sell|target|price)\s+[A-Z]{1,5}',
|
| 135 |
+
r'\$\d+(?:\.\d{2})?', # Dollar amounts
|
| 136 |
+
r'\b(?:bullish|bearish|buy|sell|hold)\b',
|
| 137 |
+
]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 138 |
|
| 139 |
+
# Look for company names and stock mentions
|
| 140 |
+
companies = re.findall(r'\b[A-Z][a-z]+(?:\s+[A-Z][a-z]+)*(?:\s+(?:Inc|Corp|Company|Ltd)\.?)?', text)
|
| 141 |
+
symbols = re.findall(r'\b[A-Z]{2,5}\b', text)
|
| 142 |
+
prices = re.findall(r'\$\d+(?:\.\d{2})?', text)
|
| 143 |
+
actions = re.findall(r'\b(?:buy|sell|hold|bullish|bearish|target|stop\s+loss)\b', text, re.IGNORECASE)
|
| 144 |
|
| 145 |
+
# Format the extracted information
|
| 146 |
+
result = "=== EXTRACTED STOCK INFORMATION ===\n\n"
|
| 147 |
|
| 148 |
if companies:
|
| 149 |
+
result += f"π Mentioned Companies: {', '.join(set(companies[:10]))}\n\n"
|
|
|
|
|
|
|
| 150 |
|
| 151 |
+
if symbols:
|
| 152 |
+
result += f"π€ Potential Stock Symbols: {', '.join(set(symbols[:10]))}\n\n"
|
| 153 |
|
| 154 |
+
if prices:
|
| 155 |
+
result += f"π° Price Mentions: {', '.join(set(prices[:10]))}\n\n"
|
| 156 |
|
| 157 |
if actions:
|
| 158 |
+
result += f"π Trading Actions: {', '.join(set(actions[:10]))}\n\n"
|
| 159 |
|
| 160 |
+
# Look for specific recommendation patterns
|
|
|
|
|
|
|
|
|
|
|
|
|
| 161 |
recommendations = []
|
| 162 |
+
sentences = text.split('.')
|
| 163 |
for sentence in sentences:
|
| 164 |
+
if any(word in sentence.lower() for word in ['buy', 'sell', 'target', 'recommendation']):
|
| 165 |
+
if any(symbol in sentence for symbol in symbols[:5]):
|
| 166 |
+
recommendations.append(sentence.strip())
|
|
|
|
| 167 |
|
| 168 |
if recommendations:
|
| 169 |
+
result += "π― Potential Recommendations:\n"
|
| 170 |
+
for rec in recommendations[:5]:
|
| 171 |
+
result += f"β’ {rec}\n"
|
|
|
|
| 172 |
|
| 173 |
+
if not any([companies, symbols, prices, actions]):
|
| 174 |
+
result += "β οΈ No clear stock recommendations found in the transcript.\n"
|
| 175 |
+
result += "This might be because:\n"
|
| 176 |
+
result += "β’ The video doesn't contain stock recommendations\n"
|
| 177 |
+
result += "β’ The audio quality was poor\n"
|
| 178 |
+
result += "β’ The content is not in English\n"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 179 |
|
| 180 |
return result
|
| 181 |
|
| 182 |
except Exception as e:
|
| 183 |
+
return f"Error extracting stock info: {str(e)}"
|
| 184 |
|
| 185 |
def cleanup_file(file_path):
|
| 186 |
"""Clean up temporary files"""
|
|
|
|
| 195 |
except:
|
| 196 |
pass
|
| 197 |
|
| 198 |
+
def system_test():
|
| 199 |
+
"""Test system components"""
|
| 200 |
+
results = []
|
| 201 |
+
|
| 202 |
+
# Test yt-dlp
|
| 203 |
+
if YT_DLP_AVAILABLE:
|
| 204 |
+
results.append("β
yt-dlp: Available")
|
| 205 |
+
try:
|
| 206 |
+
ydl = YoutubeDL({'quiet': True})
|
| 207 |
+
results.append("β
yt-dlp: Can create YoutubeDL instance")
|
| 208 |
+
except Exception as e:
|
| 209 |
+
results.append(f"β yt-dlp: Cannot create instance - {e}")
|
| 210 |
+
else:
|
| 211 |
+
results.append("β yt-dlp: Not available")
|
| 212 |
+
|
| 213 |
+
# Test Whisper
|
| 214 |
+
if WHISPER_AVAILABLE:
|
| 215 |
+
results.append("β
Whisper: Available (Type: openai-whisper)")
|
| 216 |
+
try:
|
| 217 |
+
import whisper
|
| 218 |
+
results.append("β
Whisper: OpenAI Whisper can be imported")
|
| 219 |
+
except Exception as e:
|
| 220 |
+
results.append(f"β Whisper: Cannot import - {e}")
|
| 221 |
+
else:
|
| 222 |
+
results.append("β Whisper: Not available")
|
| 223 |
|
| 224 |
+
# Test file operations
|
| 225 |
try:
|
| 226 |
+
temp_file = tempfile.NamedTemporaryFile(delete=False)
|
| 227 |
+
temp_file.write(b"test")
|
| 228 |
+
temp_file.close()
|
| 229 |
+
os.remove(temp_file.name)
|
| 230 |
+
results.append("β
File operations: Working")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 231 |
except Exception as e:
|
| 232 |
+
results.append(f"β File operations: Failed - {e}")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 233 |
|
| 234 |
+
# Test cookies
|
| 235 |
+
cookies_path = get_cookies_path()
|
| 236 |
+
if cookies_path:
|
| 237 |
+
results.append(f"β
Cookies: Found at {cookies_path}")
|
| 238 |
+
else:
|
| 239 |
+
results.append("β οΈ Cookies: Not found (may cause bot detection issues)")
|
| 240 |
|
| 241 |
+
return "\n".join(results)
|
| 242 |
|
| 243 |
+
def process_video(url, progress=gr.Progress()):
|
| 244 |
+
"""Main function to process YouTube video"""
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 245 |
|
| 246 |
# Check if required packages are available
|
| 247 |
if not YT_DLP_AVAILABLE:
|
| 248 |
+
return "Error: yt-dlp is not installed properly. Please check the requirements.", ""
|
|
|
|
|
|
|
|
|
|
|
|
|
| 249 |
|
| 250 |
if not WHISPER_AVAILABLE:
|
| 251 |
+
return "Error: OpenAI Whisper is not installed properly. Please check the requirements.", ""
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 252 |
|
| 253 |
+
if not url or not url.strip():
|
| 254 |
+
return "Please provide a valid YouTube URL", ""
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 255 |
|
| 256 |
audio_path = None
|
|
|
|
|
|
|
| 257 |
try:
|
| 258 |
+
# Validate URL
|
| 259 |
+
if not any(domain in url.lower() for domain in ['youtube.com', 'youtu.be']):
|
| 260 |
+
return "Please provide a valid YouTube URL", ""
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 261 |
|
| 262 |
# Download audio
|
| 263 |
+
progress(0.1, desc="Downloading audio...")
|
| 264 |
+
audio_path = download_audio(url)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 265 |
|
| 266 |
# Transcribe audio
|
| 267 |
+
progress(0.5, desc="Transcribing audio...")
|
|
|
|
| 268 |
transcript = transcribe_audio(audio_path)
|
|
|
|
| 269 |
|
| 270 |
if not transcript.strip():
|
| 271 |
+
return "No speech detected in the video", ""
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 272 |
|
| 273 |
# Extract stock information
|
| 274 |
+
progress(0.8, desc="Extracting stock information...")
|
| 275 |
+
stock_details = extract_stock_info_simple(transcript)
|
|
|
|
|
|
|
|
|
|
|
|
|
| 276 |
|
| 277 |
+
progress(1.0, desc="Complete!")
|
| 278 |
+
return transcript, stock_details
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 279 |
|
| 280 |
except Exception as e:
|
| 281 |
+
error_msg = f"Error processing video: {str(e)}"
|
| 282 |
+
return error_msg, ""
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 283 |
|
| 284 |
finally:
|
| 285 |
# Clean up temporary files
|
| 286 |
+
cleanup_file(audio_path)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 287 |
|
| 288 |
+
# Create Gradio interface
|
| 289 |
with gr.Blocks(
|
| 290 |
+
title="Stock Recommendation Extractor",
|
| 291 |
theme=gr.themes.Soft(),
|
| 292 |
css="""
|
| 293 |
.gradio-container {
|
| 294 |
max-width: 1200px;
|
| 295 |
margin: auto;
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 296 |
}
|
| 297 |
"""
|
| 298 |
) as demo:
|
| 299 |
|
| 300 |
gr.Markdown("""
|
| 301 |
+
# π Stock Recommendation Extractor from YouTube
|
| 302 |
|
| 303 |
+
Extract stock recommendations and trading information from YouTube videos using AI transcription.
|
| 304 |
|
| 305 |
+
**How it works:**
|
| 306 |
+
1. Downloads audio from YouTube video
|
| 307 |
+
2. Transcribes using OpenAI Whisper
|
| 308 |
+
3. Extracts stock-related information
|
|
|
|
| 309 |
|
| 310 |
+
**β οΈ Disclaimer:** This is for educational purposes only. Always do your own research!
|
| 311 |
""")
|
| 312 |
|
| 313 |
+
# Add system test section
|
| 314 |
+
with gr.Accordion("π§ͺ System Status", open=False):
|
| 315 |
+
system_status = gr.Textbox(
|
| 316 |
+
value=system_test(),
|
| 317 |
+
label="System Test Results",
|
| 318 |
+
lines=10,
|
| 319 |
+
interactive=False
|
| 320 |
+
)
|
| 321 |
+
test_btn = gr.Button("π Re-run System Test")
|
| 322 |
+
test_btn.click(fn=system_test, outputs=system_status)
|
| 323 |
+
|
| 324 |
with gr.Row():
|
| 325 |
with gr.Column(scale=1):
|
| 326 |
+
url_input = gr.Textbox(
|
| 327 |
+
label="πΊ YouTube URL",
|
| 328 |
+
placeholder="https://www.youtube.com/watch?v=...",
|
| 329 |
+
lines=2
|
| 330 |
+
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 331 |
|
| 332 |
+
process_btn = gr.Button(
|
| 333 |
+
"π Extract Stock Information",
|
| 334 |
+
variant="primary",
|
| 335 |
+
size="lg"
|
| 336 |
+
)
|
| 337 |
+
|
| 338 |
+
gr.Markdown("""
|
| 339 |
+
### π‘ Tips:
|
| 340 |
+
- Works best with financial YouTube channels
|
| 341 |
+
- Ensure video has clear audio
|
| 342 |
+
- English content works best
|
| 343 |
+
- If you get bot detection errors, try updating cookies.txt
|
| 344 |
+
""")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 345 |
|
|
|
|
| 346 |
with gr.Row():
|
| 347 |
with gr.Column():
|
| 348 |
transcript_output = gr.Textbox(
|
| 349 |
label="π Full Transcript",
|
| 350 |
+
lines=15,
|
| 351 |
+
max_lines=20,
|
| 352 |
+
show_copy_button=True
|
|
|
|
| 353 |
)
|
| 354 |
|
| 355 |
with gr.Column():
|
| 356 |
stock_info_output = gr.Textbox(
|
| 357 |
label="π Extracted Stock Information",
|
| 358 |
+
lines=15,
|
| 359 |
+
max_lines=20,
|
| 360 |
+
show_copy_button=True
|
|
|
|
| 361 |
)
|
| 362 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 363 |
# Event handlers
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 364 |
process_btn.click(
|
| 365 |
fn=process_video,
|
| 366 |
+
inputs=[url_input],
|
| 367 |
+
outputs=[transcript_output, stock_info_output],
|
| 368 |
show_progress=True
|
| 369 |
)
|
| 370 |
|
| 371 |
+
# Example section
|
| 372 |
+
gr.Markdown("### π Example URLs (Replace with actual financial videos)")
|
| 373 |
+
gr.Examples(
|
| 374 |
+
examples=[
|
| 375 |
+
["https://www.youtube.com/watch?v=dQw4w9WgXcQ"],
|
| 376 |
+
],
|
| 377 |
+
inputs=[url_input],
|
| 378 |
+
label="Click to try example"
|
| 379 |
+
)
|
| 380 |
|
|
|
|
| 381 |
if __name__ == "__main__":
|
| 382 |
+
demo.launch()
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|