convert a text file to a scroling mp4 that matches length of mp3

text2scrolling_mp4.py

from PIL import Image, ImageDraw, ImageFont
import subprocess
import textwrap
from icecream import ic

#--------------------
def get_mp3_duration(mp3_path):
    cmd = [
        "ffprobe", "-v", "error",
        "-show_entries", "format=duration",
        "-of", "default=noprint_wrappers=1:nokey=1",
        mp3_path
    ]
    result = subprocess.run(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)
    try:
        duration = float(result.stdout.strip())
        ic(f"MP3 duration: {duration:.2f} seconds")
        return duration
    except ValueError:
        ic("Could not extract duration from mp3")
        return 5.0

#--------------------
def text_to_long_image(text_file_path, output_image_path, font_path=None, font_size=24, max_line_width=1000, line_spacing=8):
    ic("Generating long image from text")
    with open(text_file_path, 'r', encoding='utf-8') as f:
        text = f.read()
    ic(f"Text length: {len(text)} characters")

    if not font_path:
        font_path = "/usr/share/fonts/truetype/dejavu/DejaVuSansMono.ttf"
    font = ImageFont.truetype(font_path, font_size)

    lines = []
    for paragraph in text.split('\n'):
        if paragraph.strip() == '':
            lines.append('')
            continue
        wrapped = textwrap.wrap(paragraph, width=100)
        for line in wrapped:
            while font.getlength(line) > max_line_width:
                cut = len(line)
                low, high = 0, cut
                while low < high:
                    mid = (low + high) // 2
                    if font.getlength(line[:mid]) <= max_line_width:
                        low = mid + 1
                    else:
                        high = mid
                cut = low - 1 if low > 0 else 1
                lines.append(line[:cut])
                line = line[cut:]
            lines.append(line)

    line_height = font.getbbox("A")[3] + line_spacing
    img_height = line_height * len(lines) + 20
    img_width = max_line_width + 40

    ic(f"Image size: {img_width}x{img_height}")

    img = Image.new("RGB", (img_width, img_height), color="white")
    draw = ImageDraw.Draw(img)

    y = 10
    for line in lines:
        draw.text((20, y), line, fill="black", font=font)
        y += line_height

    img.save(output_image_path)
    ic(f"Saved long image: {output_image_path}")

#--------------------

def scroll_image_to_video(image_path, output_video_path, mp3_path, width=512, height=768, scroll_speed_factor=1.0):
    img = Image.open(image_path)
    img_width, img_height = img.size
    ic(f"Image dimensions: {img_width}x{img_height}")

    scroll_distance = img_height - height
    ic(f"Scroll distance: {scroll_distance} pixels")

    duration = get_mp3_duration(mp3_path)

    if scroll_distance <= 0:
        ic("No scrolling needed")
        scroll_filter = f"scale={width}:{height}"
    else:
        scroll_speed = (scroll_distance / duration) * scroll_speed_factor
        ic(f"Scroll speed: {scroll_speed:.2f} px/sec")
        scroll_filter = (
            f"scale={width}:-1,"
            f"crop={width}:{height}:0:y='min(t*{scroll_speed},{scroll_distance})'"
        )

    cmd = [
        "ffmpeg",
        "-y",
        "-loop", "1",
        "-i", image_path,
        "-i", mp3_path,
        "-vf", scroll_filter,
        "-t", f"{duration}",
        "-c:v", "libx264",
        "-pix_fmt", "yuv420p",
        "-c:a", "aac",
        "-b:a", "192k",
        "-shortest",
        "-movflags", "+faststart",
        output_video_path
    ]

    ic("Running ffmpeg command:")
    ic(' '.join(cmd))
    subprocess.run(cmd, check=True)
    ic(f"Video created: {output_video_path}")
#--------------------
if __name__ == "__main__":
    text_file = "short.txt"
    long_image = "long_text.png"
    output_video = "scroll_synced.mp4"
    mp3_audio = "short.mp3"

    # Generate the long image from the text file
    text_to_long_image(text_file, long_image)

    # Create the scrolling video synchronized with the MP3 file
    #scroll_image_to_video(long_image, output_video, mp3_audio)
    # scroll_speed_factor adjust scrolling speed + 1 is faster example: +1.3 | -float is slower example: -.5
    scroll_image_to_video(long_image, output_video, mp3_audio, scroll_speed_factor=.5)