import os import json import urllib.request from PIL import Image from gtts import gTTS import cv2 import moviepy.editor as mp import logging from hercai import Hercai import uuid import time import gradio as gr from typing import Tuple, List import numpy as np # Configure logging with console output logging.basicConfig( level=logging.INFO, format='[%(asctime)s] %(message)s', handlers=[ logging.FileHandler('app.log'), logging.StreamHandler() # This will print to console ] ) LOGGER = logging.getLogger(__name__) class Text2Video: def __init__(self) -> None: """Initialize the Text2Video class.""" LOGGER.info("Initializing Text2Video application...") self.herc = Hercai() LOGGER.info("Hercai API initialized successfully") def get_image(self, img_prompt: str) -> str: """Generate an image based on the provided text prompt.""" try: LOGGER.info(f"🎨 Starting image generation for prompt: {img_prompt}") # Enhanced prompt for better comic-style results comic_style_prompt = ( f"{img_prompt}, comic book style, full scene composition, " "vibrant colors, clear speech bubbles with text, " "dramatic lighting, high contrast, detailed backgrounds, " "comic book panel layout, professional illustration" ) LOGGER.info("📝 Enhanced prompt with comic style elements") LOGGER.info(f"🔄 Sending request to Hercai API...") image_result = self.herc.draw_image( model="v3", prompt=comic_style_prompt, negative_prompt="blurry, cropped, low quality, dark, gloomy" ) image_url = image_result["url"] LOGGER.info(f"✅ Image generated successfully: {image_url}") return image_url except Exception as e: LOGGER.error(f"❌ Error generating image: {str(e)}") raise def download_img_from_url(self, image_url: str, image_path: str) -> str: """Download and process image from URL.""" try: urllib.request.urlretrieve(image_url, image_path) # Image processing for consistent quality img = Image.open(image_path) target_size = (1792, 1024) img = img.resize(target_size, Image.Resampling.LANCZOS) img.save(image_path, quality=95) LOGGER.info(f"Successfully downloaded and processed image: {image_path}") return image_path except Exception as e: LOGGER.error(f"Error downloading image: {e}") raise def text_to_audio(self, img_prompt: str, audio_path: str) -> str: """Convert text to speech with enhanced quality.""" try: LOGGER.info(f"🔊 Converting text to audio: {img_prompt}") # Create audio with enhanced parameters tts = gTTS(text=img_prompt, lang='en', slow=False) LOGGER.info("📝 Audio conversion complete") # Save audio file tts.save(audio_path) LOGGER.info(f"✅ Audio saved to: {audio_path}") return audio_path except Exception as e: LOGGER.error(f"❌ Error in audio conversion: {str(e)}") raise def get_images_and_audio(self, list_prompts: List[str]) -> Tuple[List[str], List[str]]: """Process multiple prompts to generate images and audio.""" img_list = [] audio_paths = [] LOGGER.info(f"🎬 Starting batch processing of {len(list_prompts)} prompts") for idx, img_prompt in enumerate(list_prompts, 1): try: LOGGER.info(f"📍 Processing prompt {idx}/{len(list_prompts)}") # Generate unique identifier unique_id = uuid.uuid4().hex[:8] # Process image image_path = f"scene_{idx}_{unique_id}.png" img_url = self.get_image(img_prompt) image = self.download_img_from_url(img_url, image_path) img_list.append(image) # Process audio audio_path = f"audio_{idx}_{unique_id}.mp3" audio = self.text_to_audio(img_prompt, audio_path) audio_paths.append(audio) LOGGER.info(f"✅ Completed processing prompt {idx}") except Exception as e: LOGGER.error(f"❌ Error processing prompt {idx}: {str(e)}") raise return img_list, audio_paths def create_video_from_images_and_audio(self, image_files: List[str], audio_files: List[str], output_path: str) -> None: """Create final video with enhanced quality.""" try: LOGGER.info("🎥 Starting video creation process") if len(image_files) != len(audio_files): raise ValueError("Number of images and audio files don't match") video_clips = [] for idx, (image_file, audio_file) in enumerate(zip(image_files, audio_files), 1): LOGGER.info(f"🔄 Processing scene {idx}/{len(image_files)}") # Load audio and create video clip audio_clip = mp.AudioFileClip(audio_file) video_clip = mp.ImageClip(image_file).set_duration(audio_clip.duration) video_clip = video_clip.set_audio(audio_clip) video_clips.append(video_clip) LOGGER.info(f"✅ Scene {idx} processed successfully") LOGGER.info("🔄 Concatenating all scenes") final_clip = mp.concatenate_videoclips(video_clips) LOGGER.info("💾 Writing final video file") final_clip.write_videofile( output_path, codec='libx264', fps=24, audio_codec='aac', audio_bitrate='192k', preset='medium' ) LOGGER.info("✅ Video created successfully") except Exception as e: LOGGER.error(f"❌ Error in video creation: {str(e)}") raise def generate_video(self, text: str) -> str: """Main function to generate video from text.""" try: LOGGER.info("🎬 Starting video generation process") # Split text into prompts list_prompts = [sentence.strip() for sentence in text.split(",,") if sentence.strip()] LOGGER.info(f"📝 Processed {len(list_prompts)} scenes from input text") output_path = f"comic_video_{uuid.uuid4().hex[:8]}.mp4" # Generate images and audio img_list, audio_paths = self.get_images_and_audio(list_prompts) # Create final video self.create_video_from_images_and_audio(img_list, audio_paths, output_path) LOGGER.info(f"✅ Video generation completed: {output_path}") return output_path except Exception as e: LOGGER.error(f"❌ Error in video generation: {str(e)}") raise def gradio_interface(self): """Create Gradio interface.""" LOGGER.info("🌐 Initializing Gradio interface") with gr.Blocks(theme='abidlabs/dracula_revamped') as demo: gr.HTML("""

Comic Video Generator

""") with gr.Row(): input_text = gr.Textbox( label="Comic Script", placeholder="Enter your story (separate scenes with ,,)" ) with gr.Row(): generate_btn = gr.Button("🎬 Generate Video") with gr.Row(): output = gr.Video(label="Generated Comic Video") # Example Story example_txt = """once upon a time there was a village. It was a nice place to live, except for one thing. people did not like to share.,, One day a visitor came to town. 'Hello. Does anybody have food to share?' He asked. 'No', said everyone.,, That's okay', said the visitor. 'I will make stone soup for everyone'.Then he took a stone and dropped it into a giant pot,,""" gr.Examples([[example_txt]], [input_text]) generate_btn.click(self.generate_video, inputs=[input_text], outputs=[output]) LOGGER.info("✅ Gradio interface initialized") demo.launch(debug=True) if __name__ == "__main__": text2video = Text2Video() text2video.gradio_interface()