audio-post
Skill0 starsChanged 20 days agoMITView on GitHub
对真人录音或 TTS 已生成音轨做声音后期:剪取与编排、增益和基础滤波、音乐自动避让、音效混合、淡入淡出、两遍响度处理及测量。适用于配音完成后的混音,不生成配音、不代替人工审听。
In this repository13 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- asset-scoutSkill
- auto-subtitleSkill
- content-directorSkill
- daily-ai-newsSkill
- humanizerSkill
- media-qcSkill
- narrationSkill
- pptSkill
- publishSkill
- video-motionSkill
- web-captureSkill
The file
Files
声音后期
从项目已有配音、音乐和音效开始;先听源音轨,再按内容决定处理。输入必须保留,音频缺失或工具无法审听就如实标记,不以波形/响度指标代替“听过”。
按配置说明编写项目 audio.json,执行 node scripts/studio.mjs audio --project <项目>,生成 audio/mix.wav 与同名 .json 处理证据。文件名可通过 --out 指定。已知爆音、口误或严重失真优先回源修复;滤波和响度归一不会恢复丢失的信息。
音乐按人声信号压低,参数按实际听感调整;转场音效服务于信息,不逐镜强加。正式音轨确定后更新分镜/剪辑时间与字幕,配音语速变化不靠平均字数摊时。
先识别底噪、低频嗡声、喷麦、混响或削波的具体问题,再选处理;低切只清理低频,响度归一会同时抬高噪声。稳定底噪可按配置启用轻度频谱降噪,默认关闭,不给每段干净录音套同一强度。先选包含正常说话、句尾与停顿的代表段,保留不加 BGM、同等人声音量的处理前后对照;听是否有水声、金属感、吞字或呼吸抽动,再决定全片参数。不能实际听音时可交测量与待审试听,不能宣布听感通过;明显混响、失真或丢字不承诺靠降噪恢复。
渲染器使用处理后的音轨作为显式输入;把 audio.json、实际音轨及来源随 recipe 归档,最终混音显式列入 deliverables.json。参数、源文件或混音变化后,重新检查受影响字幕与音画同步。成片视听检查交 media-qc;工具结果始终保留待审听状态。
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

