video2.py

from moviepy.editor import *
from PIL import Image
import pytesseract
import numpy as np
from gtts import gTTS
from mutagen.mp3 import MP3
import os
import uuid
import moviepy.config as mpc

# Set FFmpeg path explicitly
os.environ["IMAGEIO_FFMPEG_EXE"] = "/usr/bin/ffmpeg"
os.environ["TMPDIR"] = "/tmp"

# Create directories with full permissions
IMAGE_DIR = "/tmp/images"
AUDIO_DIR = "/tmp/sound"
CLIPS_DIR = "/tmp/video"

for dir_path in [IMAGE_DIR, AUDIO_DIR, CLIPS_DIR]:
    os.makedirs(dir_path, exist_ok=True)
    os.chmod(dir_path, 0o777)  # Ensure full permissions

def video_func(id, lines):
    print(f"Processing video for id={id}")
    # Generate audio
    tts = gTTS(text=lines[id], lang='ta', slow=False)
    audio_name = f"audio{id}.mp3"
    audio_path = os.path.join(AUDIO_DIR, audio_name)
    print(f"Saving audio to {audio_path}")
    tts.save(audio_path)

    # Verify audio file exists
    if not os.path.exists(audio_path):
        raise FileNotFoundError(f"Audio file {audio_path} was not created")
    
    audio = MP3(audio_path)
    duration = audio.info.length
    print(f"Audio duration: {duration} seconds")

    # Set image path
    IMAGE_PATH = os.path.join(IMAGE_DIR, f"slide{id}.png")
    if not os.path.exists(IMAGE_PATH):
        raise FileNotFoundError(f"Image file {IMAGE_PATH} not found")
    print(f"Using image: {IMAGE_PATH}")

    VIDEO_DURATION = duration
    HIGHLIGHT_COLOR = (255, 255, 0)  # Yellow highlight
    HIGHLIGHT_OPACITY = 0.5  # Semi-transparent

    # OCR step
    img = Image.open(IMAGE_PATH)
    data = pytesseract.image_to_data(img, output_type=pytesseract.Output.DICT)

    # Extract words and their positions
    words = []
    for i in range(len(data['text'])):
        word = data['text'][i].strip()
        if word and int(data['conf'][i]) > 60:
            x, y, w, h = data['left'][i], data['top'][i], data['width'][i], data['height'][i]
            words.append({'text': word, 'box': (x, y, w, h)})
    print(f"Found {len(words)} words in image")

    # Base image clip
    image_clip = ImageClip(IMAGE_PATH).set_duration(VIDEO_DURATION)

    # Highlight words one by one
    n_words = len(words)
    highlight_duration = VIDEO_DURATION / n_words if n_words > 0 else VIDEO_DURATION

    highlight_clips = []
    for i, word in enumerate(words):
        x, y, w, h = word['box']
        start = i * highlight_duration
        end = start + highlight_duration
        rect = ColorClip(size=(w, h), color=HIGHLIGHT_COLOR)
        rect = rect.set_opacity(HIGHLIGHT_OPACITY).set_position((x, y)).set_start(start).set_end(end)
        highlight_clips.append(rect)

    # Final video
    final_clip = CompositeVideoClip([image_clip] + highlight_clips)
    audio = AudioFileClip(audio_path)
    final_clip = final_clip.set_audio(audio)
    
    clip_name = f"clip{id}.mp4"
    video_path = os.path.join(CLIPS_DIR, clip_name)
    print(f"Writing video to {video_path}")
    
    # Write video with verbose output for debugging
    final_clip.write_videofile(video_path, fps=24, verbose=True, logger='bar')
    print(f"Video saved to {video_path}")
    return video_path