{"data":{"slug":"buxuku-smartsub","name":"SmartSub","tagline":"A desktop subtitle tool supporting ASR and translation","github_url":"https://github.com/buxuku/SmartSub","owner":"buxuku","repo":"SmartSub","owner_avatar_url":"https://avatars.githubusercontent.com/u/7866330?v=4","primary_language":"TypeScript","stars":5253,"forks":400,"topics":["deepseek","faster-whisper","fireredasr","funasr","ollama","openai","qwen3-asr","sherpa-onnx","subtitle","translate","whisper","whisper-cpp"],"archived":false,"github_pushed_at":"2026-09-16T22:35:25+00:00","maintenance_label":"Very active","stars_delta_30d":629,"url":"https://www.graphcanon.com/tools/buxuku-smartsub","markdown_url":"https://www.graphcanon.com/tools/buxuku-smartsub.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/buxuku-smartsub","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=buxuku-smartsub","description":"视频转字幕、字幕翻译、AI 配音与声音克隆、字幕烧录——免费开源的一站式桌面工具。基于 Whisper / FunASR 等本地模型离线语音转文字，批量处理 + 全平台 GPU 加速，跨 Windows / macOS / Linux。Free, open-source desktop app to generate, translate, dub & burn video subtitles — local Whisper speech-to-text, AI dubbing & voice cloning, offline, GPU-accelerated.","homepage_url":"https://smartsub.linxiaodong.com/","license":"MIT","open_issues":50,"watchers":25,"ai_summary":"SmartSub is a comprehensive desktop application for subtitling tasks including transcription, translation, proofreading, and subtitle burn-in. It supports multiple ASR engines and GPU acceleration.","readme_excerpt":"<div align=\"center\">\n\n<img src=\"./resources/icon.png\" width=\"88\" alt=\"SmartSub\" />\n\n# 妙幕 / SmartSub\n\n**视频转字幕、字幕翻译、AI 配音与声音克隆、字幕烧录——一站式开源桌面工具**\n\n让每一帧画面都能美妙地表达\n\n<a href=\"https://trendshift.io/repositories/14079?utm_source=repository-badge&amp;utm_medium=badge&amp;utm_campaign=badge-repository-14079\" target=\"_blank\" rel=\"noopener noreferrer\"><img src=\"https://trendshift.io/api/badge/repositories/14079\" alt=\"buxuku%2FSmartSub | Trendshift\" width=\"250\" height=\"55\"/></a>\n<a href=\"https://trendshift.io/repositories/14079?utm_source=trendshift-badge&amp;utm_medium=badge&amp;utm_campaign=badge-trendshift-14079\" target=\"_blank\" rel=\"noopener noreferrer\"><img src=\"https://trendshift.io/api/badge/trendshift/repositories/14079/daily?language=TypeScript\" alt=\"buxuku%2FSmartSub | Trendshift\" width=\"250\" height=\"55\"/></a>\n\n\n\n\n\n\n[中文](README.md) | [English](README_EN.md) | [日本語](README_JA.md)\n\n[下载安装](#下载安装) · [功能特性](#功能特性) · [免费方案](#全流程免费方案) · [常见问题](#常见问题) · [更新日志](https://github.com/buxuku/SmartSub/releases)\n\n</div>\n\n\n\n## 妙幕是什么\n\n妙幕（SmartSub）是一款开源的字幕与配音工具，把「语音转文字 → 字幕翻译 → 校对润色 → TTS 配音 → 烧录合成」整条流水线装进一个桌面应用，还内置在线视频下载，粘贴 B 站 / YouTube 等平台链接即可直接取材。转写基于 whisper.cpp、sherpa-onnx 等本地模型完成，文件不出本机；支持批量处理，支持 NVIDIA / AMD / Intel / Apple Silicon 硬件加速，可在 Windows、macOS、Linux 上运行。\n\n**整条流水线可以完全免费跑通**：本地模型转写、内置免费翻译源、本地 TTS 配音（含声音克隆）、本地 ffmpeg 烧录——不需要 API Key，本地环节也没有用量限制。同时可按需接入 20 个翻译服务、9 家云端听写、6 类云端配音服务作为增强。\n\n## 它能帮你做什么\n\n| 你想做的事                       | 妙幕的做法                                                   |\n| -------------------------------- | ------------------------------------------------------------ |\n| 看无字幕的外语视频、公开课       | 拖入视频，本地转写并翻译，得到双语字幕，播放器直接挂载       |\n| 给 B 站 / YouTube 在线视频配字幕 | 粘贴链接应用内直接下载，官方字幕自动配对，无需第三方下载工具 |\n| 做视频出海 / 多语言内容          | 字幕翻译成目标语言，再用 TTS 配成外语音轨，导出成品视频      |\n| 用自己的声音给视频配音           | 录一段参考音频，零样本克隆音色，整条视频用你的声音朗读       |\n| 整理播客、网课、会议录音         | 批量转写成 SRT 字幕文件，供剪辑、检索或存档                  |\n| 给成片压制字幕                   | 校对台逐句核对后硬烧或软封装，样式所见即所得                 |\n\n## 功能特性\n\n在线视频**下载** / 本地音视频 → **转写** → **翻译** → **校对** → **配音** → **合成导出**。每一步都可以独立使用，也可以串成流水线批量执行。\n\n### 在线视频下载\n\n- 粘贴链接即可下载 B 站、YouTube 等平台的在线视频，一行一条批量下载，可从混杂文本中自动识别链接\n- yt-dlp（YouTube 及 1800+ 站点）与 lux（B 站、抖音、小红书等国内平台）双引擎，按平台自动匹配，应用内一键安装 / 更新\n- 可同时下载平台官方字幕（含自动生成字幕），任务向导中自动配对——有官方字幕就无需再转写\n- 支持导入站点 Cookie（浏览器一键提取 / cookies.txt / 手动粘贴），解锁登录后可见的高清画质与会员内容，Cookie 仅保存在本地\n- 保存目录、清晰度（最佳 / 指定档位）、并发数可调，下载完成一键衔接转写 / 翻译任务\n\n### 字幕生成（转写）\n\n- 多种视频 / 音频格式批量生成字幕，并发任务数可调\n- 8 类转写引擎逐任务切换：内置 `whisper.cpp`、`faster-whisper`、`FunASR`、`Qwen3-ASR`、`FireRedASR`、`NVIDIA Parakeet`、本地 `Whisper CLI`，以及免 GPU 的云端听写（9 家服务商）\n- 本地引擎完全离线，无需联网上传；中文场景可选 FunASR / FireRedASR，英语、欧洲语言与日语可选对应 Parakeet 模型\n- AI 字幕精修（可选）：大模型语义断句 + 批量校正——断句按语义重组且时间轴仍精确到词（连接词不吊行尾、数字不被停顿劈开），校正修同音字、去语气词、规范标点；服务商默认跟随 AI 翻译配置（本地 Ollama 零成本），失败自动回退规则断句\n- 简繁转换、自定义字幕文件名（方便不同播放器挂载识别）、可选中文字幕去标点\n\n### 字幕翻译\n\n- 20 个翻译服务：内置免费翻译（必应 / 谷歌免费接口，自动回退与限速）、百度、阿里云、腾讯、讯飞、火山引擎、豆包、小牛、DeepLX、Azure、Google，以及 Ollama（本地模型）、DeepSeek、Gemini、通义千问、SiliconFlow、Azure OpenAI、[DeerAPI](https://api.deerapi.com/register?aff=QvHM) 等大模型服务\n- 兼容任意 OpenAI 风格 API，可接入自有服务\n- 输出纯译文，或「原文 + 译文」双语字幕\n- 每个 AI 服务可在界面直接配置自定义请求参数，支持导出导入，无需改代码\n\n### 字幕校对\n\n- 内置校对台，逐句对照视频检查修改，定位精准\n- 撤销 / 重做，单条删除可恢复\n- AI 一键润色\n\n### TTS 配音与声音克隆\n\n- 独立配音工作台：一份字幕 + 可选视频，逐条语音合成并自动对齐时间轴\n- 本地引擎离线免费：Kokoro 多语 103 音色、VITS 中文 174 音色\n- 声音克隆：本地 ZipVoice 零样本克隆（一段参考音频即建即用），也支持火山引擎声音复刻 2.0、ElevenLabs 即时克隆\n- 云端服务：Edge TTS 免费档、OpenAI 兼容端点（OpenAI / 硅基流动等）、Azure Speech、火山引擎豆包、ElevenLabs、Xiaomi MiMo\n- 时间轴对齐：语速预控制、实测复核、静音间隙借用；超限行列入人工处理清单（改文案 / 单行重生成 / 接受变速）\n- 逐行试听、换音色、重新合成；背景音可静音原轨或压低原轨（ducking）\n- 输出纯音频（wav / mp3）、替换音轨、混音视频或 MKV 双音轨，可同时导出对齐后的字幕\n\n### 视频合成（字幕烧录）\n\n- 硬字幕：把字幕永久烧进画面，任何播放器都能显示\n- 软字幕：以流复制方式无损封装可切换字幕轨\n- 字体、字号、颜色、描边、阴影、九宫格位置与多种预设样式\n- 所见即所得实时预览\n\n### 隐私与硬件加速\n\n- 本地处理，文件不出本机；云端服务均为可选，首次使用有隐私确认\n- GPU 加速：NVIDIA CUDA、AMD / Intel Vulkan、Apple Core ML / Metal\n- 应用内下载加速包，无需手动安装 CUDA Toolkit；加载失败自动回退 CPU\n\n## 界面一览\n\n| 视频合成（字幕烧录）                    | 字幕校对                                       |\n| --------------------------------------- | ---------","github_created_at":"2024-05-21T09:33:49+00:00","created_at":"2026-07-15T11:16:44.883014+00:00","updated_at":"2026-09-20T05:13:39.27148+00:00","categories":[{"slug":"speech-audio","name":"Speech & Audio","url":"https://www.graphcanon.com/categories/speech-audio","markdown_url":"https://www.graphcanon.com/categories/speech-audio.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/speech-audio"}],"tags":[{"slug":"asr","name":"asr"},{"slug":"gpu-acceleration","name":"gpu acceleration"},{"slug":"subtitle","name":"subtitle"},{"slug":"transcription","name":"transcription"},{"slug":"translation","name":"translation"}],"trust":{"provenance":{"is_fork":false,"github_id":803733455,"owner_type":"User","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-09-20T05:13:36.764Z","maintenance":{"label":"Very active","score":96,"methodology":"github_public_v1","releases_90d":9,"days_since_push":3,"last_release_at":"2026-09-10T06:16:52Z","stars_delta_30d":629,"open_issues_delta_30d":7},"security_summary":{"status":"no_lockfile","scanner":null,"low_count":0,"high_count":0,"last_scan_at":"2026-07-15T11:16:46.287Z","medium_count":0,"scan_profile":"none","critical_count":0}},"capability_facts":{"mcp":{"source":"repo_scan","observed_at":"2026-09-20T05:13:37.796Z","server_manifest":false},"scan":{"source":"repo_scan","observed_at":"2026-09-20T05:13:37.796Z"},"languages":{"value":["typescript","javascript"],"source":"github.language+package.json","observed_at":"2026-09-20T05:13:37.796Z"},"license_spdx":{"value":"MIT","source":"github.license","observed_at":"2026-09-20T05:13:37.796Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":null,"constraints":null,"when_to_use":["Need multi-engine ASR support including both local and online services","Require extensive customization options in AI service parameters"],"when_not_to_use":["Looking for mobile app capabilities instead of desktop application support","Do not need the complexity of multiple engines or advanced GPU acceleration configurations"],"source":"enrich:decision_facts","observed_at":"2026-07-17T12:08:24.042Z"},"constraint_facets":null,"decision_summary":[{"label":"Adopt for","value":"SmartSub is a desktop application for subtitling tasks like transcription, translation, proofreading, and subtitle burn-in with local-first processing and GPU acceleration."}]}}