anbeime/skill / skills/video-creation-suite/video-creation-suite
video-creation-suite
Unverified自媒体创作者与影片与视频编辑当需要从零原创、二次创作或深度分析视频时,请使用此套件。通过11个智能体协同,一键生成配音、音效与BGM并合成高质量视频,或自动抽帧提取分镜提示词,全流程大幅提升视频产出效率。
scripts/compose_video.py
View on GitHub#!/usr/bin/env python3
"""
合成视频脚本
根据分镜脚本合成最终视频
"""
import os
import sys
import json
import subprocess
def composite_video_from_script(script_path: str, images_dir: str, audio_dir: str, subtitles_path: str, output_path: str):
"""
根据分镜脚本合成视频
参数:
script_path: 分镜脚本路径
images_dir: 图片目录
audio_dir: 音频目录
subtitles_path: 字幕文件路径
output_path: 输出视频路径
"""
# 读取分镜脚本
with open(script_path, 'r', encoding='utf-8') as f:
script = json.load(f)
print(f"正在合成视频: {script['title']}")
print(f"分辨率: {script['resolution'][0]}x{script['resolution'][1]}")
print(f"帧率: {script['fps']}")
print(f"总时长: {script['duration']}秒")
print(f"场景数: {len(script['scenes'])}\n")
# 计算每张图片的显示时长
scenes = script['scenes']
fps = script['fps']
resolution = script['resolution']
# 创建临时文件列表
concat_file = "/tmp/concat_list.txt"
with open(concat_file, 'w') as f:
for scene in scenes:
# 使用绝对路径
image_path = os.path.abspath(os.path.join(images_dir, scene['image']))
# 每张图片显示指定的时长
duration = scene['duration']
# 使用loop和t参数来控制时长
f.write(f"file '{image_path}'\n")
f.write(f"duration {duration}\n")
# 步骤1: 从图片序列生成视频
print("步骤1: 从图片序列生成视频...")
temp_video = "/tmp/temp_video.mp4"
cmd1 = [
"ffmpeg",
"-f", "concat",
"-safe", "0",
"-i", concat_file,
"-vf", f"fps={fps},scale={resolution[0]}:{resolution[1]}:force_original_aspect_ratio=decrease,pad={resolution[0]}:{resolution[1]}:(ow-iw)/2:(oh-ih)/2",
"-c:v", "libx264",
"-preset", "ultrafast",
"-pix_fmt", "yuv420p",
"-y",
temp_video
]
try:
result = subprocess.run(cmd1, capture_output=True, text=True, timeout=300)
if result.returncode != 0:
print(f"✗ 步骤1失败")
print(f"错误: {result.stderr}")
raise Exception(f"FFmpeg步骤1失败: {result.stderr}")
print("✓ 步骤1完成\n")
except subprocess.TimeoutExpired:
print("✗ 步骤1超时")
raise Exception("FFmpeg步骤1超时")
except Exception as e:
print(f"✗ 步骤1失败: {str(e)}")
raise
# 步骤2: 添加音频
print("步骤2: 添加音频...")
# 合并背景音乐和旁白
temp_audio = "/tmp/temp_audio.mp3"
bgm_path = os.path.join(audio_dir, "background_music.mp3")
narration_path = os.path.join(audio_dir, "narration.mp3")
if os.path.exists(bgm_path) and os.path.exists(narration_path):
# 混合两个音频
cmd2a = [
"ffmpeg",
"-i", bgm_path,
"-i", narration_path,
"-filter_complex", "[0:a]volume=0.3[a0];[1:a]volume=1.0[a1];[a0][a1]amix=inputs=2:duration=first",
"-y",
temp_audio
]
result = subprocess.run(cmd2a, capture_output=True, text=True, timeout=60)
if result.returncode != 0:
print(f"⚠️ 音频混合失败,仅使用旁白: {result.stderr}")
temp_audio = narration_path
elif os.path.exists(narration_path):
temp_audio = narration_path
else:
temp_audio = None
# 步骤3: 合并视频和音频
print("步骤3: 合并视频和音频...")
os.makedirs(os.path.dirname(output_path), exist_ok=True)
if temp_audio:
cmd3 = [
"ffmpeg",
"-i", temp_video,
"-i", temp_audio,
"-c:v", "copy",
"-c:a", "aac",
"-shortest",
"-y",
output_path
]
else:
cmd3 = [
"ffmpeg",
"-i", temp_video,
"-c:v", "copy",
"-y",
output_path
]
try:
result = subprocess.run(cmd3, capture_output=True, text=True, timeout=300)
if result.returncode != 0:
print(f"✗ 步骤3失败")
print(f"错误: {result.stderr}")
raise Exception(f"FFmpeg步骤3失败: {result.stderr}")
print("✓ 步骤3完成\n")
except subprocess.TimeoutExpired:
print("✗ 步骤3超时")
raise Exception("FFmpeg步骤3超时")
except Exception as e:
print(f"✗ 步骤3失败: {str(e)}")
raise
# 步骤4: 添加字幕(可选)
if subtitles_path and os.path.exists(subtitles_path):
print("步骤4: 添加字幕...")
final_output = output_path.replace(".mp4", "_subtitled.mp4")
cmd4 = [
"ffmpeg",
"-i", output_path,
"-vf", f"subtitles='{subtitles_path}':force_style='Fontsize=24,PrimaryColour=&H00FFFFFF,BackColour=&H80000000,BorderStyle=1'",
"-c:a", "copy",
"-y",
final_output
]
try:
result = subprocess.run(cmd4, capture_output=True, text=True, timeout=300)
if result.returncode != 0:
print(f"⚠️ 字幕添加失败: {result.stderr}")
print("继续使用未添加字幕的视频\n")
else:
print("✓ 字幕添加完成\n")
output_path = final_output
except Exception as e:
print(f"⚠️ 字幕添加失败: {str(e)}")
print("继续使用未添加字幕的视频\n")
# 清理临时文件
try:
os.remove(concat_file)
os.remove(temp_video)
if temp_audio and temp_audio.startswith("/tmp"):
os.remove(temp_audio)
except:
pass
# 检查输出文件
if os.path.exists(output_path):
file_size = os.path.getsize(output_path)
print(f"\n✓ 视频合成成功!")
print(f"输出路径: {output_path}")
print(f"文件大小: {file_size / 1024 / 1024:.1f} MB")
else:
print(f"\n✗ 视频合成失败")
raise Exception("输出文件不存在")
if __name__ == "__main__":
script_path = "./output/script.json"
images_dir = "./output/images"
audio_dir = "./output/audio"
subtitles_path = "./output/subtitles/subtitles.srt"
output_path = "./output/final/three_body_video.mp4"
os.makedirs(os.path.dirname(output_path), exist_ok=True)
try:
composite_video_from_script(script_path, images_dir, audio_dir, subtitles_path, output_path)
except Exception as e:
print(f"\n✗ 视频合成失败: {str(e)}")
sys.exit(1)
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |