diff --git a/.github/copilot-instructions.md b/.github/copilot-instructions.md new file mode 100644 index 0000000..88bff7d --- /dev/null +++ b/.github/copilot-instructions.md @@ -0,0 +1,52 @@ +# Copilot Instructions + +## Project shape + +- The whole application lives in `index.html`. Template markup, CSS overrides, Vue state, persistence, import/export, audio processing, and video generation are all defined there. +- This is a browser-first tool for AI audiobook production. The README positions it around LLM-driven script analysis plus IndexTTS / Qwen3TTS workflows, with users supplying OpenAI-compatible LLM endpoints and TTS service URLs. +- Frontend dependencies are loaded in the page, not through a package manager: Vue 3 and Tailwind via CDN, `mp4-muxer` via CDN, and ffmpeg helpers from `vendor\ffmpeg` plus ffmpeg core wasm loaded at runtime. + +## Commands and validation + +- There is no repository build, test, or lint manifest (`package.json`, `requirements.txt`, etc. are absent). +- There is no built-in single-test command because the repository does not ship a test suite. +- Typical local run path is to open `index.html` in a browser or serve the repository as static files. +- For quick syntax validation after editing the inline Vue app, extract the main `
-{{ confirmDialog.message }}
+多角色音效合成有声书生成工具
-+ 把配置、素材库、脚本拆分、逐行精修和导出流程放进同一个浏览器工作台。 +
+音效和背景音乐可以自行使用其他的AI模型生成,或者前往在线免费音效网站下载:Pixabay在线音效库
-{{ rawAnalysisResult }}
+ class="bg-slate-800 text-slate-200 p-3 sm:p-4 rounded-lg text-[11px] sm:text-xs overflow-x-auto whitespace-pre-wrap max-h-[16rem] sm:max-h-[20rem] xl:max-h-[26rem]">{{ rawAnalysisResult }}
+
本网站由sdsds222制作,点击此处访问个人主页
@@ -1392,10 +2445,37 @@ 3. Qwen 音色生成参考音频