Skip to content

Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.

docs(site): refresh ecosystem evidence (#3473)

zhifu gao1e6ec0cbe2c7
diff --git a/docs/superpowers/plans/2026-08-05-ecosystem-evidence-refresh.md b/docs/superpowers/plans/2026-08-05-ecosystem-evidence-refresh.md
new file mode 100644
index 000000000..4f64160c9
--- /dev/null
+++ b/docs/superpowers/plans/2026-08-05-ecosystem-evidence-refresh.md
@@ -0,0 +1,97 @@
+# Ecosystem Evidence Refresh Implementation Plan
+
+> **For agentic workers:** REQUIRED SUB-SKILL: Use superpowers:subagent-driven-development (recommended) or superpowers:executing-plans to implement this plan task-by-task. Steps use checkbox (`- [ ]`) syntax for tracking.
+
+**Goal:** Refresh the bilingual FunASR ecosystem page with current owned-repository scale, the latest stable FunClip release, and the merged audio.cpp Fun-ASR-Nano runtime.
+
+**Architecture:** Keep the existing legacy ecosystem pages as the source surface consumed by the approved product-site builder. Add output-level assertions before editing content, update only the two bilingual pages, then refresh their existing snapshot hashes in the legacy manifest.
+
+**Tech Stack:** Static HTML, Python 3.11+, Beautiful Soup, pytest, the existing product-site builder and validator, Playwright.
+
+## Global Constraints
+
+- Follow the approved Option A product deployment hub design in `docs/superpowers/specs/2026-07-26-funasr-product-deployment-hub-design.md`.
+- Do not claim an integration is merged or released without exact public evidence.
+- Keep Chinese and English ecosystem cards structurally equivalent.
+- Preserve all existing routes, navigation, legacy content, and attributed `/go/funclip` repository link.
+- Do not add dependencies or change unrelated site content.
+
+---
+
+### Task 1: Lock the ecosystem evidence contract
+
+**Files:**
+- Modify: `web-pages/product-site/tests/test_output.py`
+
+**Interfaces:**
+- Consumes: the existing `built_site` fixture and `read_soup()` helper.
+- Produces: a bilingual regression gate for the ecosystem scale, FunClip release, and audio.cpp integration evidence.
+
+- [x] **Step 1: Write the failing output test**
+
+Add a parametrized test for `ecosystem.html` and `en/ecosystem.html`. It must assert that the rendered page contains `36K+`, links to `https://github.com/modelscope/FunClip/releases/tag/v2.1.1`, and contains an audio.cpp card linking to all of:
+
+```text
+https://github.com/0xShug0/audio.cpp
+https://github.com/0xShug0/audio.cpp/pull/155
+https://github.com/0xShug0/audio.cpp/blob/1778b23a5f6a4951c788e4bb0e7baa04f20012a2/docs/models/fun_asr_nano.md
+```
+
+The card text must include `Fun-ASR-Nano`, `CPU`, `CUDA`, `CLI`, and `OpenAI`.
+
+- [x] **Step 2: Run the test and verify RED**
+
+Run:
+
+```bash
+python -m pytest web-pages/product-site/tests/test_output.py -k ecosystem_refresh -q
+```
+
+Expected: fail because the current page says `35K+`, links FunClip `v2.1.0`, and has no audio.cpp card.
+
+### Task 2: Refresh the bilingual ecosystem source
+
+**Files:**
+- Modify: `web-pages/product-site/legacy/ecosystem.html`
+- Modify: `web-pages/product-site/legacy/en/ecosystem.html`
+- Modify: `web-pages/product-site/content/legacy-manifest.json`
+
+**Interfaces:**
+- Consumes: the output contract from Task 1 and public GitHub evidence for FunClip `v2.1.1` and audio.cpp PR #155.
+- Produces: bilingual legacy pages accepted by the unchanged product-site builder.
+
+- [x] **Step 1: Apply the minimal bilingual content update**
+
+Change the aggregate ecosystem figure from `35K+` to `36K+`, update the FunClip release link and label from `v2.1.0` to `v2.1.1`, and add one audio.cpp card under cross-platform inference. Describe only the merged native Fun-ASR-Nano implementation: pure C++/GGML, CPU/CUDA, CLI, and OpenAI-compatible local HTTP. Link the repository, merged PR, and pinned merged guide.
+
+- [x] **Step 2: Refresh only the two snapshot hashes**
+
+Compute SHA-256 for the two modified legacy pages and replace only their existing values in `legacy-manifest.json`. Keep `captured` and every unrelated file hash unchanged.
+
+- [x] **Step 3: Run the focused test and verify GREEN**
+
+Run:
+
+```bash
+python -m pytest web-pages/product-site/tests/test_output.py -k ecosystem_refresh -q
+```
+
+Expected: both language cases pass.
+
+- [x] **Step 4: Run complete site verification**
+
+Run:
+
+```bash
+python -m pytest web-pages/product-site/tests -q
+python web-pages/product-site/build.py --output /tmp/funasr-site-ecosystem-refresh
+python web-pages/product-site/validate.py /tmp/funasr-site-ecosystem-refresh
+cd web-pages/product-site/tests/browser && npm ci && npx playwright test
+git diff --check
+```
+
+Expected: all tests and validators pass, Playwright desktop/mobile coverage passes, and the diff is whitespace-clean.
+
+- [ ] **Step 5: Commit and publish for review**
+
+Create a signed+DCO commit containing exactly the test, two pages, manifest hashes, and this plan. Push `codex/site-ecosystem-refresh-20260805`, open a ready PR, wait for required checks, and merge only after the exact pushed head is green and mergeable. After deployment, run `python scripts/check_funasr_website_static.py` and direct HTTPS checks against both ecosystem routes.
diff --git a/web-pages/product-site/content/legacy-manifest.json b/web-pages/product-site/content/legacy-manifest.json
index aec2ea0ce..89d5a9537 100644
--- a/web-pages/product-site/content/legacy-manifest.json
+++ b/web-pages/product-site/content/legacy-manifest.json
@@ -39,7 +39,7 @@
"css/index.1682178c.css": "b5f6034ce886c25773928c114e4290d1c491a3d081677b488810198efd76b4e1",
"decoder.js": "f2023a28036b1ff58e5bdabbb72ea427f8faaf60e89acd78ff183c313bdb0a2c",
"donors.html": "eec6b789b29c27d372925ae0ad3e13d4c6e62978b3c38c4da9e3aec129f10934",
- "ecosystem.html": "1defb2b8de27bbb85290813f8f9263b96fe5b9c67c821165645fca26d52d19b6",
+ "ecosystem.html": "1007c2bd0cd6fffb849aa8aa651c077f9c4f52d0c171b8ffb1137c47c4bfe9e4",
"en/blog/cantonese-speech-recognition.html": "b9bd146198206f781c37f341c370a690dc83cd7163334e34b4e7f65e69af8ff2",
"en/blog/chinese-speech-recognition.html": "16b68bd2a1122236c580c62275002ae2c0875ca14e32686e88d3b5ccfa9f56ef",
"en/blog/fun-asr-nano-guide.html": "1746c708bd0606188de4266d8255984dd9e14634af33c54ad65a5f05b767be1b",
@@ -71,7 +71,7 @@
"en/blog/voice-activity-detection-python.html": "d12f5d0f4ae07c3434f6188d12f5603b8abb9578413b79133668638d1c5e18a7",
"en/blog/which-funasr-model.html": "1207180aa929a3aa51db136c1098f4edc8f07405e2b56fe033998bf996f8ec7f",
"en/donors.html": "eacb7f9434b1b29e13e9b0175bb9680556ee72ab4dbd155e68f9c912eb0212b1",
- "en/ecosystem.html": "3b7dc72e8925d60716e144ea8771ca433275cf1281ba29cb872d15fc7451e3f2",
+ "en/ecosystem.html": "c3e4f9e007953799a090565af0a7bcceaf66ae645c6d42591675080fb2d1a7dc",
"en/index.html": "4950bc2e0b337637300c5bcd45d1820d9636c4f1dab34e0ec8d3443be44bd277",
"en/llama-cpp.html": "dbfbd125ebbf07f7d923c91b122818f8b09484ce54c1e80eba66a89d22dfb5b4",
"en/models.html": "c2add621edc3aa63e6d49587f089ee5ac6c05faf5f68e14fef70a7a5ab443a4d",
diff --git a/web-pages/product-site/legacy/ecosystem.html b/web-pages/product-site/legacy/ecosystem.html
index a10d5a0f2..ee439ca21 100644
--- a/web-pages/product-site/legacy/ecosystem.html
+++ b/web-pages/product-site/legacy/ecosystem.html
@@ -75,7 +75,7 @@ footer a{color:#60a5fa}
<div class="stat-bar">
<div class="stat"><div class="stat-num">50+</div><div class="stat-label">集成项目</div></div>
<div class="stat"><div class="stat-num">50+</div><div class="stat-label">支持语种</div></div>
-<div class="stat"><div class="stat-num">35K+</div><div class="stat-label">GitHub Stars</div></div>
+<div class="stat"><div class="stat-num">36K+</div><div class="stat-label">GitHub Stars</div></div>
<div class="stat"><div class="stat-num">100万+</div><div class="stat-label">月安装量</div></div>
</div>
@@ -84,7 +84,7 @@ footer a{color:#60a5fa}
<div class="card">
<div class="card-title"><a href="/go/funclip">FunClip</a></div>
<div class="card-stars">6K+ stars</div>
-<div class="card-desc">基于语音识别的智能视频剪辑。自动生成字幕,支持按关键词和说话人剪辑视频片段。<a href="https://github.com/modelscope/FunClip/releases/tag/v2.1.0">下载 v2.1.0</a></div>
+<div class="card-desc">基于语音识别的智能视频剪辑。自动生成字幕,支持按关键词和说话人剪辑视频片段。<a href="https://github.com/modelscope/FunClip/releases/tag/v2.1.1">下载 v2.1.1</a></div>
<div><span class="card-tag">视频</span><span class="card-tag green">官方</span></div>
</div>
<div class="card">
@@ -348,6 +348,12 @@ footer a{color:#60a5fa}
<div><span class="card-tag">移动端</span><span class="card-tag">边缘计算</span></div>
</div>
<div class="card">
+<div class="card-title"><a href="https://github.com/0xShug0/audio.cpp">audio.cpp</a></div>
+<div class="card-stars">1.1K+ stars</div>
+<div class="card-desc">已合并的 <a href="https://github.com/0xShug0/audio.cpp/pull/155">Fun-ASR-Nano 原生实现</a>通过纯 C++ / GGML 在 CPU 或 CUDA 上运行,提供 CLI 与兼容 OpenAI 的本地 HTTP 转写服务。<a href="https://github.com/0xShug0/audio.cpp/blob/1778b23a5f6a4951c788e4bb0e7baa04f20012a2/docs/models/fun_asr_nano.md">查看固定提交的使用与验证指南</a>。</div>
+<div><span class="card-tag">Fun-ASR-Nano</span><span class="card-tag">C++ / GGML</span></div>
+</div>
+<div class="card">
<div class="card-title"><a href="https://github.com/RapidAI/RapidASR">RapidASR</a></div>
<div class="card-stars">608 stars</div>
<div class="card-desc">跨平台 ASR 推理库,基于 ONNX Runtime 和 FunASR。开箱即用,支持中英文混合识别。</div>
diff --git a/web-pages/product-site/legacy/en/ecosystem.html b/web-pages/product-site/legacy/en/ecosystem.html
index 42ea10840..390e4687e 100644
--- a/web-pages/product-site/legacy/en/ecosystem.html
+++ b/web-pages/product-site/legacy/en/ecosystem.html
@@ -76,7 +76,7 @@ footer a{color:#60a5fa}
<div class="stat-bar">
<div class="stat"><div class="stat-num">50+</div><div class="stat-label">Integrations</div></div>
<div class="stat"><div class="stat-num">50+</div><div class="stat-label">Languages</div></div>
-<div class="stat"><div class="stat-num">35K+</div><div class="stat-label">GitHub Stars</div></div>
+<div class="stat"><div class="stat-num">36K+</div><div class="stat-label">GitHub Stars</div></div>
<div class="stat"><div class="stat-num">1M+</div><div class="stat-label">pip installs/month</div></div>
</div>
@@ -85,7 +85,7 @@ footer a{color:#60a5fa}
<div class="card">
<div class="card-title"><a href="/go/funclip">FunClip</a></div>
<div class="card-stars">6K+ stars</div>
-<div class="card-desc">ASR-powered video clipping. Automatic subtitles, keyword and speaker-based clip extraction. <a href="https://github.com/modelscope/FunClip/releases/tag/v2.1.0">Download v2.1.0</a></div>
+<div class="card-desc">ASR-powered video clipping. Automatic subtitles, keyword and speaker-based clip extraction. <a href="https://github.com/modelscope/FunClip/releases/tag/v2.1.1">Download v2.1.1</a></div>
<div><span class="card-tag">Video</span><span class="card-tag green">Official</span></div>
</div>
<div class="card">
@@ -349,6 +349,12 @@ footer a{color:#60a5fa}
<div><span class="card-tag">Mobile</span><span class="card-tag">Edge</span></div>
</div>
<div class="card">
+<div class="card-title"><a href="https://github.com/0xShug0/audio.cpp">audio.cpp</a></div>
+<div class="card-stars">1.1K+ stars</div>
+<div class="card-desc">The merged <a href="https://github.com/0xShug0/audio.cpp/pull/155">native Fun-ASR-Nano implementation</a> runs with pure C++ and GGML on CPU or CUDA, exposing a CLI and local OpenAI-compatible HTTP transcription service. <a href="https://github.com/0xShug0/audio.cpp/blob/1778b23a5f6a4951c788e4bb0e7baa04f20012a2/docs/models/fun_asr_nano.md">Read the pinned usage and verification guide</a>.</div>
+<div><span class="card-tag">Fun-ASR-Nano</span><span class="card-tag">C++ / GGML</span></div>
+</div>
+<div class="card">
<div class="card-title"><a href="https://github.com/RapidAI/RapidASR">RapidASR</a></div>
<div class="card-stars">608 stars</div>
<div class="card-desc">Cross-platform ASR inference library based on ONNX Runtime and FunASR. Ready to use, supports Chinese-English mixed recognition.</div>
diff --git a/web-pages/product-site/tests/test_output.py b/web-pages/product-site/tests/test_output.py
index 2d463b4d3..748e9fb37 100644
--- a/web-pages/product-site/tests/test_output.py
+++ b/web-pages/product-site/tests/test_output.py
@@ -229,6 +229,34 @@ def test_ecosystem_surfaces_released_subtitle_edit_funasr_backends(
assert marker in text
[email protected]('relative', ('ecosystem.html', 'en/ecosystem.html'))
+def test_ecosystem_refresh_tracks_current_release_and_merged_native_runtime(
+ built_site, relative
+):
+ soup = read_soup(built_site / relative)
+ text = soup.get_text(' ', strip=True)
+
+ assert '36K+' in text
+ assert soup.select_one(
+ 'a[href="https://github.com/modelscope/FunClip/releases/tag/v2.1.1"]'
+ )
+
+ anchor = soup.select_one(
+ '.card-title a[href="https://github.com/0xShug0/audio.cpp"]'
+ )
+ assert anchor
+ card = anchor.find_parent(class_='card')
+ assert card
+ links = {link.get('href') for link in card.select('a[href]')}
+ assert {
+ 'https://github.com/0xShug0/audio.cpp/pull/155',
+ 'https://github.com/0xShug0/audio.cpp/blob/1778b23a5f6a4951c788e4bb0e7baa04f20012a2/docs/models/fun_asr_nano.md',
+ } <= links
+ card_text = card.get_text(' ', strip=True)
+ for marker in ('Fun-ASR-Nano', 'CPU', 'CUDA', 'CLI', 'OpenAI'):
+ assert marker in card_text
+
+
@pytest.mark.parametrize(
('relative', 'peer', 'markers'),
(