mp4tomp4withaudioandsubs

import os
from moviepy.editor import VideoFileClip, CompositeVideoClip, TextClip, concatenate_audioclips, AudioFileClip
from gtts import gTTS
import subprocess  # For running ffmpeg command

# Define paths
video_path = "/Users/p/Downloads/mango_from_seed_to_superfood.mp4"
output_video_path = "/Users/p/Downloads/mango_from_seed_to_superfood_final2.mp4"
final_audio_path = "/Users/p/Downloads/mango_from_seed_to_superfood_final_audio2.mp3"
temp_audio_dir = '/Users/p/Downloads/temp_audio'  # Temporary audio files directory
os.makedirs(temp_audio_dir, exist_ok=True)  # Ensure the directory exists

# Load the video
video = VideoFileClip(video_path)

# Define subtitles and their timings with manual line breaks
subtitles = [
    ("Did you know a single mango tree\ncan produce up to 4,000 fruits?", 0, 5),
    ("Originating from the Indo-Burma region,\nmangoes are a gift of ancient forests.", 6, 12),
    ("India leads global production, with\nsignificant volumes from China and Thailand.", 13, 18),
    ("Rich in vitamins A and C,\nmangoes are nutritional powerhouses.", 19, 24),
    ("In India, the heartland of mangoes,\ndiverse varieties thrive.", 25, 30),
    ("The cost of cultivating one acre of mango\nis approximately Rs. 2,00,000.", 31, 36),
    ("With an annual yield of 2.5 to 3 tonnes\nper acre, farmers can expect a profit\nof around Rs. 70,000.", 37, 42),
    ("The real magic begins with transformation.\nFrom purees to pickles, mangoes offer\nendless possibilities.", 43, 48),
    ("Value addition unlocks a world of\neconomic opportunity.", 49, 54),
    ("Dive into the vibrant world of mangoes and\nexplore their untapped potential.", 55, 60)
]

# Function to generate text-to-speech and create a corresponding audio clip
def text_to_speech(text, filename):
    tts = gTTS(text=text.replace('\n', ' '), lang='en')  # Replace newlines for TTS
    tts.save(filename)
    return AudioFileClip(filename)

composite_clips = [video]  # Start with the main video
audio_clips = []

# Create audio and text clips
for idx, (text, start, end) in enumerate(subtitles):
    audio_filename = os.path.join(temp_audio_dir, f'audio{idx}.mp3')
    tts_audio = text_to_speech(text, audio_filename)
    audio_clips.append(tts_audio.set_start(start))

    # Create text clip with adjusted styling
    subtitle_clip = TextClip(text, fontsize=50, color='white', font='Arial-Bold', align='center', stroke_color='black', stroke_width=1)
    subtitle_clip = subtitle_clip.set_start(start).set_duration(end - start).set_position(('center', 'center'))
    composite_clips.append(subtitle_clip)

# Combine the video with text clips
final_video = CompositeVideoClip(composite_clips)

# Combine all audio clips and use them as the video's audio
final_audio = concatenate_audioclips(audio_clips)
final_video = final_video.set_audio(final_audio)

# Write the video with subtitles to file
final_video.write_videofile(output_video_path, codec="libx264", fps=video.fps)

# Save the final combined audio separately
final_audio.write_audiofile(final_audio_path)

# Use ffmpeg to merge the new audio with the new video
ffmpeg_command = [
    'ffmpeg',
    '-i', output_video_path,
    '-i', final_audio_path,
    '-c:v', 'copy',
    '-c:a', 'aac',
    '-strict', 'experimental',
    output_video_path.replace('.mp4', '_with_audio.mp4')
]
subprocess.run(ffmpeg_command, check=True)

# Clean up temporary audio files
for file in os.listdir(temp_audio_dir):
    os.remove(os.path.join(temp_audio_dir, file))
os.rmdir(temp_audio_dir)