python moviepy 图文批量合成带字幕口播视频

最近在研究将图片和文本批量合成为带字幕口播视频

主要是基于python的moviepy库

from generator import audio, pics, subs, video


def main():
    texts_input = 'example'
    pics_input = 'example'

    # 图片分辨率预处理
    pics.adjust(pics_input)

    # 文字转语音
    audio.text_to_audio(texts_input)

    # 语音转视频
    video.audio_to_video(texts_input, pics_input)

    # 生成字幕
    subs.download_subs(texts_input)

    # 生成字幕视频
    video.attach_subs(texts_input)


if __name__ == "__main__":
    main()

以下是核心的图片+文本转视频逻辑

import json
from mutagen.mp3 import MP3
from moviepy import editor
from moviepy.editor import VideoFileClip, TextClip, CompositeVideoClip
import generator.pics as pics


def audio_to_video(text_input, pics_input):
    audio_filepath = './output/audios/' + text_input.replace(' ', '_') + '.mp3'
    video_filepath = './outpu