新媒体视频的工业化生产：从脚本到分发的技术闭环

import subprocess, pathlib

def to_vertical(src, dst):
    vf = (
        "split=2[bg][fg];"
        "[bg]scale=1080:1920:force_original_aspect_ratio=increase,"
        "crop=1080:1920,boxblur=20:5[bg];"
        "[fg]scale=1080:-2[fg];"
        "[bg][fg]overlay=(W-w)/2:(H-h)/2"
    )
    cmd = ["ffmpeg", "-y", "-i", str(src), "-filter_complex", vf,
           "-c:v", "libx264", "-preset", "fast", "-crf", "23",
           "-c:a", "aac", "-b:a", "128k",
           "-movflags", "+faststart", str(dst)]
    subprocess.run(cmd, check=True)

for f in pathlib.Path("input").glob("*.mp4"):
    to_vertical(f, pathlib.Path("output") / f.name)

import whisper

def format_time(s):
    h = int(s // 3600); m = int((s % 3600) // 60); sec = s % 60
    return f"{h:02d}:{m:02d}:{sec:06.3f}".replace(".", ",")

model = whisper.load_model("base")
result = model.transcribe("output/demo.mp4", language="zh")
with open("output/demo.srt", "w", encoding="utf-8") as f:
    for i, seg in enumerate(result["segments"], 1):
        f.write(f"{i}\n{format_time(seg['start'])} --> {format_time(seg['end'])}\n"
                f"{seg['text'].strip()}\n\n")

subprocess.run([
    "ffmpeg", "-y", "-i", "output/demo.mp4",
    "-vf", "subtitles=output/demo.srt:force_style='FontSize=18,PrimaryColour=&HFFFFFF&'",
    "-c:v", "libx264", "-crf", "20", "-c:a", "copy",
    "output/demo_sub.mp4"
], check=True)

subprocess.run([
    "ffmpeg", "-y", "-ss", "3", "-i", "output/demo_sub.mp4",
    "-vframes", "1", "-q:v", "2", "output/cover.jpg"
], check=True)

subprocess.run([
    "ffmpeg", "-y", "-i", "output/demo_sub.mp4",
    "-af", "loudnorm=I=-14:TP=-1.5:LRA=11",
    "-c:v", "copy", "output/final.mp4"
], check=True)