Spaces:
Sleeping
Sleeping
File size: 22,022 Bytes
ff099b9 f24f171 5a6a643 6ecc602 5140a89 c907607 f24f171 4c33be9 f24f171 4c33be9 f24f171 5a6a643 f24f171 8f90878 f24f171 c00b679 e575ced c00b679 e575ced c00b679 f24f171 c00b679 1a68cc7 f24f171 5a6a643 c907607 f24f171 c907607 f24f171 c907607 f24f171 5a6c7a7 f24f171 5a6c7a7 f24f171 c907607 f24f171 454b9fc f24f171 c907607 4c33be9 f24f171 4c33be9 f24f171 4c33be9 f24f171 4c33be9 f24f171 e575ced 4c33be9 e575ced f24f171 c907607 f24f171 4c33be9 c907607 f24f171 5a6a643 f24f171 5a6a643 f24f171 5a6a643 f24f171 c907607 f24f171 5a6a643 c907607 f24f171 c907607 f24f171 5a6a643 f24f171 c907607 f24f171 5a6a643 f24f171 c907607 f24f171 c907607 5a6a643 c907607 5a6a643 c907607 f24f171 c907607 5a6a643 f24f171 5a6a643 f24f171 5a6a643 f24f171 c907607 5a6a643 c907607 f24f171 c907607 f24f171 c907607 5a6a643 c907607 5a6a643 c907607 f24f171 c907607 f24f171 5a6a643 c907607 f24f171 c907607 f24f171 c907607 c00b679 d61861b c00b679 d61861b c00b679 d61861b c00b679 d61861b f24f171 5a6a643 f24f171 5a6a643 f24f171 5a6a643 f24f171 5a6a643 f24f171 5a6a643 f24f171 5a6a643 c907607 f24f171 c907607 f24f171 c907607 f24f171 c907607 f24f171 c907607 5a6a643 f24f171 c00b679 f24f171 c00b679 f24f171 c00b679 48c897c c00b679 5140a89 f24f171 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 |
import streamlit as st
from pathlib import Path
import torch
from transformers import pipeline, AutoModelForCausalLM, AutoTokenizer
from PIL import Image, ImageDraw, ImageFont
import tempfile
import os
from moviepy.editor import *
import numpy as np
from gtts import gTTS
import textwrap
from concurrent.futures import ThreadPoolExecutor
import io
import unicodedata
import re
import requests
import random
import logging
import time
from typing import Optional, List, Dict, Tuple
from bs4 import BeautifulSoup
import requests
from io import BytesIO
class ImageScraper:
def __init__(self):
self.headers = {
'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'
}
def scrape_pexels(self, query: str) -> List[str]:
urls = []
try:
url = f"https://www.pexels.com/search/{query.replace(' ', '%20')}/"
response = requests.get(url, headers=self.headers)
soup = BeautifulSoup(response.text, 'html.parser')
for img in soup.find_all('img', src=True):
if 'photos' in img['src'] and 'pexels.com' in img['src']:
urls.append(img['src'])
except Exception as e:
print(f"Pexels scraping error: {e}")
return urls
def scrape_unsplash(self, query: str) -> List[str]:
urls = []
try:
url = f"https://unsplash.com/s/photos/{query.replace(' ', '-')}"
response = requests.get(url, headers=self.headers)
soup = BeautifulSoup(response.text, 'html.parser')
for img in soup.find_all('img', src=True):
if 'images.unsplash.com' in img['src']:
urls.append(img['src'])
except Exception as e:
print(f"Unsplash scraping error: {e}")
return urls
def get_images(self, query: str, num_images: int = 15) -> List[str]:
all_urls = []
all_urls.extend(self.scrape_pexels(query))
all_urls.extend(self.scrape_unsplash(query))
# Remove duplicates and limit to num_images
return list(set(all_urls))[:num_images]
class EnhancedVideoGenerator:
def __init__(self):
try:
self.setup_logging()
self.setup_device()
self.initialize_models()
self.setup_workspace()
self.load_assets()
self.setup_themes()
self.image_scraper = ImageScraper()
except Exception as e:
logging.error(f"Initialization failed: {str(e)}")
raise RuntimeError("Failed to initialize video generator")
def setup_logging(self):
"""Configure logging for the application"""
logging.basicConfig(
level=logging.INFO,
format='%(asctime)s - %(name)s - %(levelname)s - %(message)s',
handlers=[
logging.FileHandler('video_generator.log'),
logging.StreamHandler()
]
)
self.logger = logging.getLogger(__name__)
def setup_device(self):
"""Set up computing device (CPU/GPU)"""
self.device = "cuda" if torch.cuda.is_available() else "cpu"
self.logger.info(f"Using device: {self.device}")
def initialize_models(self):
"""Initialize all AI models"""
try:
# Text generation model initialization with error handling
try:
self.text_generator = pipeline(
'text-generation',
model='gpt2',
device=0 if self.device == "cuda" else -1
)
except Exception as e:
self.logger.warning(f"Text generator initialization failed: {str(e)}")
self.text_generator = None
# Skip the StableDiffusion model initialization as it requires additional setup
self.image_model = None
# Initialize stability API attribute
self.stability_api = None
except Exception as e:
self.logger.error(f"Model initialization failed: {str(e)}")
# Don't raise exception, allow initialization with degraded functionality
pass
def setup_workspace(self):
"""Set up working directory and resources"""
self.temp_dir = Path(tempfile.mkdtemp())
self.asset_dir = self.temp_dir / "assets"
self.asset_dir.mkdir(exist_ok=True)
def setup_themes(self):
"""Set up visual themes"""
self.themes = {
'Professional': {
'bg': (240, 240, 240),
'accent': (0, 120, 212),
'text': (33, 33, 33)
},
'Creative': {
'bg': (255, 250, 240),
'accent': (255, 123, 0),
'text': (51, 51, 51)
},
'Educational': {
'bg': (248, 249, 250),
'accent': (40, 167, 69),
'text': (33, 37, 41)
}
}
def load_assets(self):
"""Load visual assets and fonts"""
try:
# Try multiple font options
font_options = [
"arial.ttf",
"/usr/share/fonts/truetype/dejavu/DejaVuSans.ttf",
"/System/Library/Fonts/Helvetica.ttc"
]
for font_path in font_options:
try:
self.font = ImageFont.truetype(font_path, 40)
break
except OSError:
continue
else:
self.font = ImageFont.load_default()
self.logger.warning("Using default font - custom font loading failed")
except Exception as e:
self.logger.error(f"Asset loading failed: {str(e)}")
def generate_visual_assets(self, script: str, style: str) -> List[Dict]:
"""Generate relevant visual assets based on script content"""
try:
# Extract key topics from script
topics = self.extract_key_topics(script)
assets = []
for topic in topics:
# Generate AI image
image = self.generate_ai_image(topic, style)
if image:
assets.append({
'type': 'image',
'data': image,
'topic': topic
})
return assets
except Exception as e:
self.logger.error(f"Visual asset generation failed: {str(e)}")
return []
def create_enhanced_frame(
self,
text: str,
theme: dict,
frame_number: int,
total_frames: int,
background_image: Optional[Image.Image] = None,
size: Tuple[int, int] = (1920, 1080) # Upgraded to 1080p
) -> np.ndarray:
"""Create a visually enhanced frame with background, text, and effects"""
try:
# Create base frame
if background_image:
# Resize and crop background to fit
bg = background_image.resize(size, Image.LANCZOS)
frame = np.array(bg)
else:
frame = np.full((size[1], size[0], 3), theme['bg'], dtype=np.uint8)
# Convert to PIL Image for drawing
img = Image.fromarray(frame)
draw = ImageDraw.Draw(img, 'RGBA')
# Add subtle gradient overlay
overlay = Image.new('RGBA', size, (0, 0, 0, 0))
overlay_draw = ImageDraw.Draw(overlay)
overlay_draw.rectangle(
[0, 0, size[0], size[1]],
fill=(255, 255, 255, 100) # Semi-transparent white
)
img = Image.alpha_composite(img.convert('RGBA'), overlay)
# Add text with improved styling
text = self.clean_text(text)
wrapped_text = textwrap.fill(text, width=50)
# Calculate text position
text_bbox = draw.textbbox((0, 0), wrapped_text, font=self.font)
text_width = text_bbox[2] - text_bbox[0]
text_height = text_bbox[3] - text_bbox[1]
text_x = (size[0] - text_width) // 2
text_y = size[1] - text_height - 100 # Position at bottom
# Draw text background
padding = 20
draw.rectangle(
[
text_x - padding,
text_y - padding,
text_x + text_width + padding,
text_y + text_height + padding
],
fill=(0, 0, 0, 160) # Semi-transparent black
)
# Draw text
draw.text(
(text_x, text_y),
wrapped_text,
fill=(255, 255, 255, 255),
font=self.font
)
# Add progress bar with animation
self.draw_animated_progress_bar(
draw,
frame_number,
total_frames,
size,
theme
)
return np.array(img)
except Exception as e:
self.logger.error(f"Frame creation failed: {str(e)}")
# Return fallback frame
return np.full((size[1], size[0], 3), theme['bg'], dtype=np.uint8)
def draw_animated_progress_bar(
self,
draw: ImageDraw.Draw,
frame_number: int,
total_frames: int,
size: Tuple[int, int],
theme: dict
):
"""Draw an animated progress bar with effects"""
try:
progress = frame_number / total_frames
bar_width = int(size[0] * 0.8) # 80% of screen width
bar_height = 6
x_offset = (size[0] - bar_width) // 2
y_position = size[1] - 40
# Draw background bar
draw.rectangle(
[x_offset, y_position, x_offset + bar_width, y_position + bar_height],
fill=(200, 200, 200, 160)
)
# Draw progress with gradient effect
progress_width = int(bar_width * progress)
for x in range(progress_width):
alpha = int(255 * (x / bar_width)) # Gradient effect
draw.line(
[x_offset + x, y_position, x_offset + x, y_position + bar_height],
fill=(theme['accent'][0], theme['accent'][1], theme['accent'][2], alpha)
)
# Add animated highlight
highlight_pos = x_offset + progress_width
if highlight_pos < x_offset + bar_width:
draw.rectangle(
[highlight_pos-2, y_position-1, highlight_pos+2, y_position + bar_height+1],
fill=(255, 255, 255, 200)
)
except Exception as e:
self.logger.error(f"Progress bar drawing failed: {str(e)}")
def generate_voice_over(self, script: str) -> AudioFileClip:
"""Generate voice-over audio using gTTS"""
try:
audio_path = self.temp_dir / "voice.mp3"
tts = gTTS(
text=script,
lang='en',
slow=False
)
tts.save(str(audio_path))
return AudioFileClip(str(audio_path))
except Exception as e:
self.logger.error(f"Voice-over generation failed: {str(e)}")
return AudioFileClip(duration=len(script.split()) * 0.3)
def create_video(self, script: str, style: str, duration: int, output_path: str, selected_images: List[str]) -> str:
"""Create video with selected images"""
try:
# Progress bar
progress_bar = st.progress(0)
status_text = st.empty()
# Generate voice-over (20%)
status_text.text("Creating voice-over...")
audio = self.generate_voice_over(script)
progress_bar.progress(20)
# Process selected images (40%)
status_text.text("Processing images...")
processed_images = []
for img_url in selected_images:
response = requests.get(img_url)
img = Image.open(BytesIO(response.content))
img = img.resize((1920, 1080), Image.Resampling.LANCZOS)
processed_images.append(np.array(img))
progress_bar.progress(40)
# Create frames with transitions
fps = 30
total_frames = int(duration * fps)
frames = []
status_text.text("Generating frames...")
frames_per_image = total_frames // len(processed_images)
for idx, img in enumerate(processed_images):
for _ in range(frames_per_image):
frames.append(img)
# Add transition frames
if idx < len(processed_images) - 1:
next_img = processed_images[idx + 1]
for alpha in np.linspace(0, 1, 15):
transition_frame = (1 - alpha) * img + alpha * next_img
frames.append(transition_frame.astype(np.uint8))
progress_bar.progress(70)
# Create video clip
status_text.text("Compiling video...")
video = ImageSequenceClip(frames, fps=fps)
video = video.set_audio(audio)
progress_bar.progress(90)
# Write final video
status_text.text("Saving video...")
video.write_videofile(
output_path,
fps=fps,
codec='libx264',
audio_codec='aac',
threads=4,
preset='ultrafast'
)
progress_bar.progress(100)
status_text.text("Video generation complete!")
return output_path
except Exception as e:
self.logger.error(f"Video creation failed: {str(e)}")
raise
def generate_visual_assets(self, script: str, style: str) -> List[Dict]:
"""Generate relevant visual assets based on script content"""
try:
# Simplified asset generation for faster processing
topics = self.extract_key_topics(script)[:3] # Limit to 3 topics
assets = []
for topic in topics:
# Create simple colored backgrounds instead of AI images
img = Image.new('RGB', (1920, 1080), self.themes[style]['bg'])
assets.append({
'type': 'image',
'data': img,
'topic': topic
})
return assets
except Exception as e:
self.logger.error(f"Visual asset generation failed: {str(e)}")
return []
@staticmethod
def clean_text(text: str) -> str:
"""Clean and normalize text for display"""
if not isinstance(text, str):
text = str(text)
# Normalize unicode characters
text = unicodedata.normalize('NFKD', text)
# Remove non-ASCII characters
text = text.encode('ascii', 'ignore').decode('ascii')
# Replace problematic characters
replacements = {
'β': '-', # en dash
'β': '-', # em dash
'"': '"', # smart quotes
'"': '"', # smart quotes
''': "'", # smart apostrophe
''': "'", # smart apostrophe
'β¦': '...', # ellipsis
}
for old, new in replacements.items():
text = text.replace(old, new)
# Remove any remaining non-standard characters
text = re.sub(r'[^\x00-\x7F]+', '', text)
return text.strip()
def extract_key_topics(self, script: str) -> List[str]:
"""Extract main topics from the script for visual asset generation"""
try:
# Simple keyword extraction based on noun phrases
# In a production environment, you might want to use a proper NLP library
sentences = script.split('.')
topics = []
for sentence in sentences:
words = sentence.strip().split()
if len(words) >= 2:
# Extract potential noun phrases (pairs of words)
topics.append(' '.join(words[:2]))
# Remove duplicates and limit to top 5 topics
return list(dict.fromkeys(topics))[:5]
except Exception as e:
self.logger.error(f"Topic extraction failed: {str(e)}")
return ["default topic"]
def generate_ai_image(self, prompt: str, style: str) -> Optional[Image.Image]:
"""Generate an AI image using Stability AI"""
try:
if not self.stability_api:
return None
# Enhance prompt based on style
style_prompts = {
'Professional': "professional, corporate, clean, modern",
'Creative': "artistic, vibrant, innovative, dynamic",
'Educational': "clear, informative, academic, detailed"
}
enhanced_prompt = f"{prompt}, {style_prompts.get(style, '')}, high quality, 4k"
# Generate image
response = self.stability_api.generate(
prompt=enhanced_prompt,
samples=1,
width=1920,
height=1080
)
if response and len(response) > 0:
image_data = response[0].image
return Image.open(io.BytesIO(image_data))
return None
except Exception as e:
self.logger.error(f"AI image generation failed: {str(e)}")
return None
def cleanup(self):
"""Clean up temporary files and resources"""
try:
for file in self.temp_dir.glob('*'):
try:
if file.is_file():
file.unlink()
elif file.is_dir():
import shutil
shutil.rmtree(file)
except Exception as e:
self.logger.warning(f"Failed to delete {file}: {str(e)}")
self.temp_dir.rmdir()
except Exception as e:
self.logger.error(f"Cleanup failed: {str(e)}")
def __enter__(self):
return self
def __exit__(self, exc_type, exc_val, exc_tb):
self.cleanup()
# Streamlit UI Class
class VideoGeneratorUI:
def __init__(self):
self.generator = EnhancedVideoGenerator()
self.setup_ui()
def setup_ui(self):
st.title("Enhanced Video Generator")
# Step 1: Input prompt
prompt = st.text_input("Enter your video topic/prompt")
if prompt:
# Step 2: Image Selection
st.subheader("Select Images for Your Video")
images = self.generator.image_scraper.get_images(prompt)
if not images:
st.warning("No images found. Try a different search term.")
return
selected_images = []
cols = st.columns(3)
for idx, img_url in enumerate(images):
with cols[idx % 3]:
try:
response = requests.get(img_url)
img = Image.open(BytesIO(response.content))
st.image(img, use_column_width=True)
if st.checkbox(f"Select Image {idx + 1}", key=f"img_{idx}"):
selected_images.append(img_url)
except:
continue
# Step 3: Video Generation (only show if images are selected)
if selected_images:
st.subheader("Video Generation Settings")
col1, col2 = st.columns(2)
with col1:
style = st.selectbox("Choose style", options=list(self.generator.themes.keys()))
with col2:
duration = st.slider("Video duration (seconds)", 10, 300, 60, 10)
if st.button("Generate Video"):
try:
output_path = os.path.join(os.getcwd(), f"generated_video_{int(time.time())}.mp4")
video_path = self.generator.create_video(prompt, style, duration, output_path, selected_images)
if os.path.exists(video_path):
st.success("Video generated successfully!")
st.video(video_path)
with open(video_path, 'rb') as video_file:
st.download_button(
"Download Video",
video_file.read(),
file_name=os.path.basename(video_path),
mime="video/mp4"
)
except Exception as e:
st.error(f"Failed to generate video: {str(e)}")
if __name__ == "__main__":
ui = VideoGeneratorUI() |