利用 AI 大模型和自动化工作流,根据主题或关键词一键生成高清短视频。Generate HD short videos from a topic or keyword with an automated AI workflow.
Python ai-video-generatorcontent-creationffmpeginstagram-reelsllmpythonshort-videosubtitlestext-to-speechtiktokvideo-automationvideo-workflowworkflow-automationyoutube-shorts
from concurrent.futures import ThreadPoolExecutor
import shutil
import subprocess
import threading
import wave
import numpy as np
from app.models.schema import VideoAspect, VideoConcatMode
from app.services import video
def test_concurrent_combinations_own_intermediate_clips(tmp_path, monkeypatch):
binary = video.utils.get_ffmpeg_binary()
audio = tmp_path / "audio.wav"
with wave.open(str(audio), "wb") as writer:
writer.setnchannels(1)
writer.setsampwidth(2)
writer.setframerate(24000)
writer.writeframes(b"\0\0" * 4800)
sources = []
for color in ("red", "blue"):
source = tmp_path / f"{color}.mp4"
subprocess.run([binary, "-y", "-f", "lavfi", "-i", f"color=c={color}:s=64x64:r=10", "-t", "0.4", "-c:v", "libx264", "-pix_fmt", "yuv420p", str(source)], check=True, capture_output=True)
sources.append(source)
monkeypatch.setattr(VideoAspect, "to_resolution", lambda self: (64, 64))
monkeypatch.setattr(video, "_get_configured_video_codec", lambda: "libx264")
monkeypatch.setattr(video, "_get_effective_video_codec", lambda *args: "libx264")
# Serialize the two real encodes, then hold both until they have completed.
# This deterministically exposes a shared filename before either concat.
encode_lock = threading.Lock()
both_encoded = threading.Barrier(2)
real_write = video._write_videofile_with_codec_fallback
paths = []
def encode(clip, target, **kwargs):
with encode_lock:
paths.append(target)
real_write(clip, target, threads=1, **kwargs)
both_encoded.wait(timeout=15)
monkeypatch.setattr(video, "_write_videofile_with_codec_fallback", encode)
foreign = tmp_path / "temp-clip-1.mp4"
shutil.copyfile(sources[0], foreign)
sentinel = foreign.read_bytes()
def combine(index):
output = tmp_path / f"output-{index}.mp4"
video.combine_videos(str(output), [str(sources[index])], str(audio), video_concat_mode=VideoConcatMode.sequential, threads=1)
return output
with ThreadPoolExecutor(max_workers=2) as executor:
outputs = list(executor.map(combine, range(2)))
for index, output in enumerate(outputs):
with video.VideoFileClip(str(output), audio=False) as clip:
pixel = np.mean(clip.get_frame(0), axis=(0, 1))
channel = 0 if index == 0 else 2
assert pixel[channel] > 200
assert pixel[2 if index == 0 else 0] < 30
assert len(set(paths)) == 2
assert foreign.read_bytes() == sentinel
assert sorted(p.name for p in tmp_path.iterdir()) == ["audio.wav", "blue.mp4", "output-0.mp4", "output-1.mp4", "red.mp4", "temp-clip-1.mp4"]