docs(site): publish FunClip MOSS clipping guide

web-pages/product-site/content/legacy-manifest.json

@@ -21,8 +21,9 @@
     "blog/funasr-vs-faster-whisper-chinese.html": "bfe9bb8017be80c7e7f4587726f43f6064f1dc4c65e39788f836fdfb0c9789f7",
     "blog/funasr-vs-whisper-benchmark.html": "b7b49adf24d20570abb09b733ce03d4a50a4a0e98e746b4a9320f453e01cce84",
     "blog/funclip-v2-1-0-video-clipping-release.html": "88f6c44e5332d1746c4db0fc97d755ef82f12e46d8f1c1c0ad9351152ff9dfc1",
+    "blog/funclip-v2-2-0-moss-speaker-clipping.html": "7127fb3d6493500d1e44e2168427b3e9adf9ec9bcd840d9da8fd5a1323af9709",
     "blog/generate-subtitles-srt-vtt-from-audio-video.html": "f1133235673d441cc654310581f59a319f182314ab7c0b0824af87da9f4a0591",
-    "blog/index.html": "fbfa2611c97cc5426d78450078df36fb1818da1a5eeaef718e6a18fa9fd3d287",
+    "blog/index.html": "0698ce3d08c4b61c1e1e3f94142d3564581d162765081437229d2c4f3fff5ff7",
     "blog/japanese-speech-recognition.html": "399f5ce84e68ac00854bdf70b52b1cde6795efca7c800fd6492035a37a7c1b68",
     "blog/lightweight-speech-recognition-cpu.html": "d6270066222ed5baef0df108e6a4155f4a87bb5482169f23f218794d371f3281",
     "blog/punctuation-restoration-python.html": "6cd26e03bf75b4343afd7b351c2681ebd513b058fb549c5888802be9ff3a1229",
@@ -59,8 +60,9 @@
     "en/blog/funasr-vs-faster-whisper-chinese.html": "abf31d3827dfba9b31ccd4e76229de9e4bfff00ab8bcee33bd718c88249630f4",
     "en/blog/funasr-vs-whisper-benchmark.html": "367d4a8a1cc09ac932c80cdad065127925c5e5f942a806ace683c16dc1132769",
     "en/blog/funclip-v2-1-0-video-clipping-release.html": "229baf59adf2c3290541d9b3c8a6243992406ba84b712714e9d14599205cb1d4",
+    "en/blog/funclip-v2-2-0-moss-speaker-clipping.html": "6456c699b4837e5c1fc3741cda40a58a1d0caed8498351445e2dd6314dd89b3c",
     "en/blog/generate-subtitles-srt-vtt-from-audio-video.html": "1530d4b9e94820a0b60d801e8d1c19b7aeb031b2f2b76c092657392e47706c6d",
-    "en/blog/index.html": "526253689a12ead8058ef9f45b9cdc37254f6507ab6d23511a742d88fddb95c2",
+    "en/blog/index.html": "fc46e28078b49173d9f124afc51bed070b94b3738850b077a7fffcdaa965bc3a",
     "en/blog/japanese-speech-recognition.html": "c476adcc2be1ed7c19e2345cc91b8ee6a0e04efd794b9b6476dd5d9b3a2c8b04",
     "en/blog/lightweight-speech-recognition-cpu.html": "0e02c0e0853b12613d32c250f593c05c2052f18231207b63ee01d6e6600a86f7",
     "en/blog/punctuation-restoration-python.html": "d6e30049f48d9dc6e6bc22eb567830013e4161eca51215616738b675dd1ffc40",
@@ -106,7 +108,7 @@
     "models.html": "724c88917c9934ffa66948cf3eb583fed930dd689d691b7d21fa22ee8332e606",
     "quickstart.html": "a543bf07c2cfc9a69ef3831f15c9e9d3e571a826a73cf7e4bdc8e9eb7b910254",
     "robots.txt": "1305b0f680b3e9f73b01301dc474423230dffab85731a379a694f30c73c07f2d",
-    "sitemap.xml": "bcc17d763cd1b0d317b3af8e52cc40fde58c4a80f3fb5664a12b5503ecf2fc19",
+    "sitemap.xml": "ed91fc0211b48458b77c56c1e600d055c013a6bb4b461821c3ec78dcff4ac26a",
     "static/liveplayer/liveplayer-component.min.js": "ea10fda35574cc3a77a3e7187095467d1a409009a2d39690051d33d0e7055d30",
     "static/liveplayer/liveplayer-lib.min.js": "ad67b4e1188c586ec218674a6db9968daf71ed4fbc262258f061c1f931d54250",
     "static/offline/index.html": "868781c621ddabe5d3f088d6b739e0c08e6cdc150587af6515cc323852018e62",

web-pages/product-site/data/deployments.json

@@ -98,6 +98,8 @@
         {"label": "merged native SGLang Omni MOSS runtime and H100 benchmark", "url": "https://github.com/sgl-project/sglang-omni/pull/914"},
         {"label": "FunASR ecosystem integration guide", "url": "https://github.com/modelscope/FunASR/blob/main/docs/moss_transcribe_diarize.md"},
         {"label": "FunASR AutoModel adapter merge and exact-head CI", "url": "https://github.com/modelscope/FunASR/pull/3558"},
+        {"label": "FunClip v2.2.0 MOSS speaker-clipping guide", "url": "https://www.funasr.com/en/blog/funclip-v2-2-0-moss-speaker-clipping.html"},
+        {"label": "FunClip v2.2.0 release and checksums", "url": "https://github.com/modelscope/FunClip/releases/tag/v2.2.0"},
         {"label": "moss-transcribe.cpp third-party C++17 and GGUF runtime", "url": "https://github.com/localai-org/moss-transcribe.cpp"},
         {"label": "LocalAI third-party moss-transcribe-cpp backend", "url": "https://github.com/mudler/LocalAI"}
       ],

web-pages/product-site/legacy/blog/funclip-v2-2-0-moss-speaker-clipping.html

@@ -0,0 +1,87 @@
+<!DOCTYPE html>
+<html lang="zh">
+<head>
+  <meta charset="utf-8">
+  <meta name="viewport" content="width=device-width,initial-scale=1">
+  <title>FunClip v2.2.0：MOSS 长音频说话人识别与视频剪辑 | FunASR</title>
+  <meta name="description" content="FunClip v2.2.0 集成第三方 MOSS-Transcribe-Diarize，通过 vLLM 完成长音频 ASR、说话人分段、SRT 和按说话人剪辑。">
+  <meta property="og:title" content="FunClip v2.2.0：MOSS 说话人识别与视频剪辑">
+  <meta property="og:description" content="一条经过端到端验证的 vLLM 路径，输出说话人分段、SRT，并支持按说话人剪辑。">
+  <meta property="og:type" content="article">
+  <meta property="og:url" content="https://www.funasr.com/blog/funclip-v2-2-0-moss-speaker-clipping.html">
+  <meta property="og:image" content="https://www.funasr.com/img/funclip-v2-1-0-interface.jpg">
+  <link rel="canonical" href="https://www.funasr.com/blog/funclip-v2-2-0-moss-speaker-clipping.html">
+  <link rel="alternate" hreflang="en" href="https://www.funasr.com/en/blog/funclip-v2-2-0-moss-speaker-clipping.html">
+  <link rel="alternate" hreflang="zh" href="https://www.funasr.com/blog/funclip-v2-2-0-moss-speaker-clipping.html">
+  <link rel="alternate" hreflang="x-default" href="https://www.funasr.com/en/blog/funclip-v2-2-0-moss-speaker-clipping.html">
+  <script type="application/ld+json">{"@context":"https://schema.org","@type":"BlogPosting","headline":"FunClip v2.2.0：MOSS 长音频说话人识别与视频剪辑","datePublished":"2026-08-31","dateModified":"2026-08-31","author":{"@type":"Organization","name":"FunASR / Tongyi Lab"},"publisher":{"@type":"Organization","name":"FunASR"},"mainEntityOfPage":{"@type":"WebPage","@id":"https://www.funasr.com/blog/funclip-v2-2-0-moss-speaker-clipping.html"},"image":"https://www.funasr.com/img/funclip-v2-1-0-interface.jpg"}</script>
+  <style>
+  p code,li code,td code,.proof code{overflow-wrap:anywhere}
+  :root{--primary:#2563eb;--accent:#047857;--surface:#f8fafc;--border:#dbe3ec;--text:#0f172a;--soft:#475569}
+  *{box-sizing:border-box}body{margin:0;font-family:Inter,-apple-system,BlinkMacSystemFont,"Segoe UI",sans-serif;color:var(--text);line-height:1.72;background:#fff}a{color:var(--primary);text-decoration:none}a:hover{text-decoration:underline}.container{max-width:840px;margin:auto;padding:0 24px}.nav{position:sticky;top:0;z-index:10;background:#fff;border-bottom:1px solid var(--border);padding:14px 0}.nav .container{max-width:1120px;display:flex;align-items:center;gap:20px}.nav-logo{font-weight:800;color:var(--text)}.nav-logo span{color:var(--primary)}.nav-links{display:flex;gap:16px;margin-left:auto}.nav-btn{padding:8px 14px;background:var(--primary);color:#fff;border-radius:7px}article{padding:64px 0 80px}h1{font-size:2.1rem;line-height:1.28;margin:0 0 12px}h2{font-size:1.35rem;margin:36px 0 12px;padding-top:20px;border-top:1px solid var(--border)}p,li{color:var(--soft)}.meta{color:#64748b}.lead{font-size:1.08rem;color:var(--text)}.hero-media{width:100%;aspect-ratio:2.04/1;object-fit:cover;border:1px solid var(--border);border-radius:8px;margin:18px 0 24px}pre{overflow:auto;background:#172033;color:#e2e8f0;padding:18px;border-radius:8px;line-height:1.6}code{font-family:ui-monospace,SFMono-Regular,Menlo,monospace}p code,li code,td code{background:var(--surface);padding:2px 5px;border-radius:4px;color:#1d4ed8}table{width:100%;border-collapse:collapse;margin:18px 0}th,td{text-align:left;vertical-align:top;padding:11px;border-bottom:1px solid var(--border)}th{background:var(--surface)}.proof{border-left:4px solid var(--accent);background:#f0fdf4;padding:14px 16px;color:#14532d}.cta{margin-top:30px;padding:22px;border:1px solid var(--border);border-radius:8px}.cta a{font-weight:700}footer{padding:28px;background:#0f172a;color:#94a3b8;text-align:center}@media(max-width:800px){.nav-links{display:none}h1{font-size:1.65rem}}
+  </style>
+</head>
+<body>
+<nav class="nav"><div class="container"><a class="nav-logo" href="/">Fun<span>ASR</span></a><div class="nav-links"><a href="/deploy/">部署</a><a href="/blog/">博客</a><a href="/ecosystem.html">生态</a></div><a href="/en/blog/funclip-v2-2-0-moss-speaker-clipping.html">EN</a><a class="nav-btn" href="https://github.com/modelscope/FunASR">GitHub</a></div></nav>
+<article><div class="container">
+  <h1>FunClip v2.2.0：用 MOSS 做长音频说话人识别与视频剪辑</h1>
+  <p class="meta">2026-08-31 · FunClip Release</p>
+  <img class="hero-media" src="/img/funclip-v2-1-0-interface.jpg" alt="FunClip 本地视频、字幕识别和智能剪辑界面">
+  <p class="lead">FunClip v2.2.0 新增一条可选的 MOSS 路径：把长音频交给 vLLM 服务，FunASR 将模型输出归一化为文本、说话人身份和时间段，FunClip 再生成 SRT 或按 <code>spkS01</code>、<code>spkS02</code> 剪辑。</p>
+  <p><a href="https://github.com/OpenMOSS/MOSS-Transcribe-Diarize">MOSS-Transcribe-Diarize</a> 是 OpenMOSS 维护的第三方模型，不属于 FunASR 或 FunClip。集成固定使用 <code>OpenMOSS-Team/MOSS-Transcribe-Diarize</code> revision <code>e8681d68e7042738ffca8ac8212bc8fcb1131ab8</code>，并明确保留模型归属与支持边界。</p>
+
+  <h2>数据路径</h2>
+  <table><thead><tr><th>阶段</th><th>职责</th></tr></thead><tbody>
+    <tr><td>vLLM</td><td>加载固定 MOSS revision，通过 <code>/v1/audio/transcriptions</code> 返回带时间与说话人标记的 JSON。</td></tr>
+    <tr><td>FunASR 1.4.9+</td><td>解析转写结果并生成统一的 <code>text</code>、<code>timestamp</code> 与 <code>sentence_info</code>。</td></tr>
+    <tr><td>FunClip 2.2.0</td><td>渲染说话人 SRT、按说话人选段，或把时间段交给现有音视频剪辑流程。</td></tr>
+  </tbody></table>
+
+  <h2>1. 启动固定 revision 的 vLLM 服务</h2>
+  <pre><code>python -m venv .venv-moss
+. .venv-moss/bin/activate
+pip install -U vllm
+
+vllm serve OpenMOSS-Team/MOSS-Transcribe-Diarize \
+  --revision e8681d68e7042738ffca8ac8212bc8fcb1131ab8 \
+  --served-model-name moss-transcribe-diarize \
+  --trust-remote-code --host 127.0.0.1 --port 8898</code></pre>
+  <p>先用真实音频检查服务契约。当前经过验证的格式是 <code>response_format=json</code>：</p>
+  <pre><code>curl -fsS http://127.0.0.1:8898/v1/audio/transcriptions \
+  -F file=@sample.wav \
+  -F model=moss-transcribe-diarize \
+  -F response_format=json \
+  -F max_completion_tokens=8192</code></pre>
+
+  <h2>2. 启动 FunClip</h2>
+  <pre><code>python -m pip install -U -r requirements.txt
+python funclip/launch.py \
+  --model moss \
+  --moss-backend vllm \
+  --moss-base-url http://127.0.0.1:8898/v1 \
+  --moss-max-tokens 8192</code></pre>
+  <p>远端服务需要 bearer token 时，把凭据放进 <code>MOSS_API_KEY</code> 环境变量；FunClip 不要求把 token 写进命令行或仓库。</p>
+
+  <h2>能力与边界</h2>
+  <ul>
+    <li>支持长音频 ASR、说话人身份、SRT，以及按说话人剪辑，包括不足一秒的有效短说话片段。</li>
+    <li>MOSS 提供段级时间戳，适合按整段或说话人剪辑；任意文本的精确字符级剪辑仍应使用 Paraformer。</li>
+    <li>全局说话人身份依赖模型看到连续音频，因此不接外部 VAD 或说话人模型，避免预切块破坏身份一致性。</li>
+    <li>如果最后一个 MOSS 片段因 token 上限没有结束时间戳，FunClip 会明确报错并提示提高 <code>--moss-max-tokens</code>，不会静默丢掉尾段。</li>
+  </ul>
+  <p>完整生产部署、健康检查和容量边界见 <a href="/deploy/moss-transcribe-diarize.html">MOSS 双语部署指南</a>。</p>
+
+  <h2>下载与校验 v2.2.0</h2>
+  <pre><code>FunClip-2.2.0.tar.gz
+SHA256 994c5d9cf392b74b36284d526eca8bada1560a3e7825ab7baa9c673a1b4ef216
+
+FunClip-2.2.0.zip
+SHA256 4f5a7d33d9ea65467f29b55b15ed5be18e64de7e57e2f9f36fe51a23b40557e7</code></pre>
+  <p>从 <a href="https://github.com/modelscope/FunClip/releases/tag/v2.2.0">FunClip v2.2.0 Release</a> 下载归档和 <code>SHA256SUMS</code>，不要从不明镜像复制二进制或凭据。</p>
+
+  <div class="proof">发布内容对应 commit <code>c205bf32a8b11226ff5e8acb9a3c7a1f00cd3b06</code>。仓库测试结果为 84 passed、1 skipped；H100 + vLLM 两说话人样例返回 S01/S02、有效 SRT，并成功按 S02 剪辑。这个验证证明发布链路可用，不代表所有语言、音频条件或并发规模。</div>
+  <div class="cta"><p>先按部署指南跑通一段真实双人音频，再进入 FunClip 处理自己的视频。</p><a href="https://github.com/modelscope/FunClip/releases/tag/v2.2.0">查看 FunClip v2.2.0 发布与下载</a></div>
+</div></article>
+<footer>© 2024-2026 Tongyi Lab, Alibaba Group</footer>
+</body>
+</html>