Skip to content

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

docs: surface MOSS in model discovery

LauraGPT944295dd0cb1
diff --git a/model_zoo/huggingface_models.md b/model_zoo/huggingface_models.md
index ad367dea5..0b5c6e283 100644
--- a/model_zoo/huggingface_models.md
+++ b/model_zoo/huggingface_models.md
@@ -117,6 +117,12 @@ Here we provided several pretrained models on different datasets. The details of
[//]: # (| [SOND](https://www.modelscope.cn/models/damo/speech_diarization_sond-en-us-callhome-8k-n16k4-pytorch/summary) | CallHome (60 hours) | 12M | Speaker diarization, profiles and records, English |)
+### Third-party Unified Transcription and Diarization
+
+| Model Name | Publisher | Parameters | Notes |
+|:----------:|:---------:|:----------:|:------|
+| [MOSS-Transcribe-Diarize](https://huggingface.co/OpenMOSS-Team/MOSS-Transcribe-Diarize) | OpenMOSS | See official model card | Third-party Apache-2.0 model. One offline request returns transcription, timestamps, and speaker labels; the FunASR integration does not require an external VAD service. See the [deployment guide](../docs/moss_transcribe_diarize.md). |
+
### Timestamp Prediction Models
[//]: # (| Model Name | Language | Training Data | Parameters | Notes |)
diff --git a/model_zoo/modelscope_models.md b/model_zoo/modelscope_models.md
index d8966b289..f5c787719 100644
--- a/model_zoo/modelscope_models.md
+++ b/model_zoo/modelscope_models.md
@@ -106,6 +106,12 @@ Here we provided several pretrained models on different datasets. The details of
| [SOND](https://www.modelscope.cn/models/damo/speech_diarization_sond-zh-cn-alimeeting-16k-n16k4-pytorch/summary) | AliMeeting (120 hours) | 40.5M | Speaker diarization, profiles and records, Chinese |
| [SOND](https://www.modelscope.cn/models/damo/speech_diarization_sond-en-us-callhome-8k-n16k4-pytorch/summary) | CallHome (60 hours) | 12M | Speaker diarization, profiles and records, English |
+### Third-party Unified Transcription and Diarization
+
+| Model Name | Publisher | Parameters | Notes |
+|:----------:|:---------:|:----------:|:------|
+| [MOSS-Transcribe-Diarize](https://huggingface.co/OpenMOSS-Team/MOSS-Transcribe-Diarize) | OpenMOSS | See official model card | Third-party Apache-2.0 model. One offline request returns transcription, timestamps, and speaker labels; FunASR integration does not require an external VAD service. See the [deployment guide](../docs/moss_transcribe_diarize.md). |
+
### Timestamp Prediction
| Model Name | Language | Training Data | Parameters | Notes |
|:--------------------------------------------------------------------------------------------------:|:--------------:|:-------------------:|:----------:|:------|
diff --git a/model_zoo/modelscope_models_zh.md b/model_zoo/modelscope_models_zh.md
index 2a2ddfa20..0626d2738 100644
--- a/model_zoo/modelscope_models_zh.md
+++ b/model_zoo/modelscope_models_zh.md
@@ -111,6 +111,12 @@
| [SOND](https://www.modelscope.cn/models/damo/speech_diarization_sond-zh-cn-alimeeting-16k-n16k4-pytorch/summary) | AliMeeting (120 小时) | 40.5M | 中文 |
| [SOND](https://www.modelscope.cn/models/damo/speech_diarization_sond-en-us-callhome-8k-n16k4-pytorch/summary) | CallHome (60 小时) | 12M | 英文 |
+### 第三方统一转写与说话人分离
+
+| 模型名字 | 发布方 | 模型参数 | 备注 |
+|:--------:|:------:|:--------:|:----|
+| [MOSS-Transcribe-Diarize](https://huggingface.co/OpenMOSS-Team/MOSS-Transcribe-Diarize) | OpenMOSS | 以官方模型卡为准 | 第三方 Apache-2.0 模型,单次非实时请求输出转写、时间戳和说话人标签;FunASR 集成无需额外拼接 VAD 服务。参见[部署指南](../docs/moss_transcribe_diarize_zh.md)。 |
+
### 时间戳预测模型
| 模型名字 | 语言 | 训练数据 | 模型参数 | 备注 |
diff --git a/web-pages/product-site/content/legacy-manifest.json b/web-pages/product-site/content/legacy-manifest.json
index bcf95ba82..81e180669 100644
--- a/web-pages/product-site/content/legacy-manifest.json
+++ b/web-pages/product-site/content/legacy-manifest.json
@@ -80,7 +80,7 @@
"en/ecosystem.html": "f02ee516440dfc3757165df801fbf93b403b2ef1fdbee1a2dee26cd2a3ef9fa3",
"en/index.html": "d86effc3bf4d218eabd465cbbe026f62c98dd50110ddd58251d9d271a64882a2",
"en/llama-cpp.html": "dbfbd125ebbf07f7d923c91b122818f8b09484ce54c1e80eba66a89d22dfb5b4",
- "en/models.html": "321a63caace107d5ce6d470948c2f5954f2d8e25306d8227b4a85ec3663ad49e",
+ "en/models.html": "87ef70a357dcd430703c290eff4319d2b4ffefcde2f854d91bc250bbd29b9fa2",
"en/quickstart.html": "7453cd7d582079eb4082373bce6e5de26492458b73c41617bbb1bd4248bda6ac",
"en/vs-whisper.html": "4833b93195957210436f791a9d8ebc726f0af93b809d740a5cee1e53de0ac390",
"favicon.ico": "e5edd9f12088fa920b848485c62c72a093f27bfba7b6a46b4e7b63c34eb5126a",
@@ -105,7 +105,7 @@
"lang-redirect.js": "e1189080a8e81eb76cf39e0538d0378b6631f5aa87e4679c0fd4fd34d1f6c197",
"llama-cpp.html": "892b5c05e3375bcc7cd0af343b629f069d8ec4f372e6470cd830cdf2181a11d5",
"logo.png": "ac0ff97f2b0203c3fad4ac5e90231d93e36d98cc8e90d383483de150b0dd0728",
- "models.html": "724c88917c9934ffa66948cf3eb583fed930dd689d691b7d21fa22ee8332e606",
+ "models.html": "01209193cd48363f00789061fbfad0a2b387582e2507ae5e875687773ef52fbc",
"quickstart.html": "a543bf07c2cfc9a69ef3831f15c9e9d3e571a826a73cf7e4bdc8e9eb7b910254",
"robots.txt": "1305b0f680b3e9f73b01301dc474423230dffab85731a379a694f30c73c07f2d",
"sitemap.xml": "ed91fc0211b48458b77c56c1e600d055c013a6bb4b461821c3ec78dcff4ac26a",
diff --git a/web-pages/product-site/legacy/en/models.html b/web-pages/product-site/legacy/en/models.html
index ba543ed47..f1341fc8d 100644
--- a/web-pages/product-site/legacy/en/models.html
+++ b/web-pages/product-site/legacy/en/models.html
@@ -3,11 +3,11 @@
<head><meta http-equiv="Cache-Control" content="no-cache, no-store, must-revalidate"><meta http-equiv="Pragma" content="no-cache"><meta http-equiv="Expires" content="0">
<meta charset="utf-8">
<meta name="viewport" content="width=device-width,initial-scale=1">
-<title>FunASR Models — SenseVoice, Paraformer, Fun-ASR-Nano, cam++</title>
-<meta name="description" content="Choose the right FunASR model: SenseVoice (Mandarin, Cantonese, English, Japanese, and Korean), Paraformer (Chinese accuracy and reliable character timestamps), Fun-ASR-Nano (LLM-based, vLLM accelerated), and cam++ (speaker diarization).">
+<title>FunASR Models — SenseVoice, Paraformer, Fun-ASR-Nano, MOSS, cam++</title>
+<meta name="description" content="Choose the right FunASR model: SenseVoice, Paraformer, Fun-ASR-Nano, the third-party MOSS transcription and diarization model, and cam++.">
<link rel="icon" href="../logo.png">
<meta property="og:title" content="FunASR Models — Speech Recognition Model Comparison">
-<meta property="og:description" content="Compare FunASR models: SenseVoice, Paraformer, Fun-ASR-Nano for speech recognition. Choose by speed, accuracy, language support, and features.">
+<meta property="og:description" content="Compare FunASR models: SenseVoice, Paraformer, Fun-ASR-Nano, and third-party MOSS. Choose by speed, accuracy, language support, and speaker-label needs.">
<meta property="og:type" content="article">
<meta property="og:url" content="https://www.funasr.com/en/models.html">
<meta property="og:image" content="https://www.funasr.com/logo.png">
@@ -79,6 +79,7 @@ footer a{color:#60a5fa}
<tr><td><strong>Fun-ASR-MLT-Nano</strong></td><td>vLLM</td><td>31</td><td>800M</td><td>Separate multilingual checkpoint</td></tr>
<tr><td><strong>SenseVoice Small</strong></td><td>Non-autoregressive; benchmark your hardware</td><td>zh, yue, en, ja, ko</td><td>234M</td><td>Five-language ASR, emotion detection</td></tr>
<tr><td><strong>Paraformer-zh</strong></td><td>Non-autoregressive; benchmark your hardware</td><td>zh, yue</td><td>220M</td><td>Best Chinese accuracy</td></tr>
+<tr><td><strong>MOSS-Transcribe-Diarize</strong></td><td>Batch; NVIDIA GPU</td><td>See official model card</td><td>Third-party</td><td>Long-form transcription + speaker labels</td></tr>
<tr><td><strong>cam++</strong></td><td>realtime</td><td>any</td><td>7.2M</td><td>Speaker diarization &amp; verification</td></tr>
</tbody>
</table>
@@ -162,6 +163,20 @@ result = model.generate(input="audio.wav")
print(result[0]["text"])</pre>
</div>
+<div class="model-card">
+<div class="model-name">MOSS-Transcribe-Diarize <span class="badge badge-orange">third-party model</span></div>
+<div class="model-meta">Published by OpenMOSS &middot; Model ID: <code>OpenMOSS-Team/MOSS-Transcribe-Diarize</code> &middot; <a href="https://huggingface.co/OpenMOSS-Team/MOSS-Transcribe-Diarize">HuggingFace</a> &middot; <a href="/en/deploy/moss-transcribe-diarize.html">FunASR deployment guide</a></div>
+<div class="model-desc">A third-party Apache-2.0 model published by OpenMOSS that produces long-form transcription, timestamps, and speaker labels in one offline request. FunASR maps structured results for AutoModel, vLLM, and SGLang Omni. It does not require an external VAD service, which does not mean the model has no internal segmentation.</div>
+<div class="model-features">
+<span class="badge badge-blue">Transcription + speaker labels</span>
+<span class="badge badge-green">No external VAD</span>
+<span class="badge badge-blue">vLLM / SGLang Omni</span>
+<span class="badge badge-orange">Non-realtime</span>
+</div>
+<h3>When to use</h3>
+<p>Long-form offline transcription for multi-speaker meetings, interviews, podcasts, and calls. This is not FunASR-owned model weight; validate overlap, timestamps, speaker consistency, and capacity on real multi-speaker audio before production.</p>
+</div>
+
<h2>Supporting Models</h2>
<div class="model-card">
diff --git a/web-pages/product-site/legacy/models.html b/web-pages/product-site/legacy/models.html
index d3ef7a1e7..b564f14f7 100644
--- a/web-pages/product-site/legacy/models.html
+++ b/web-pages/product-site/legacy/models.html
@@ -3,11 +3,11 @@
<head><meta http-equiv="Cache-Control" content="no-cache, no-store, must-revalidate"><meta http-equiv="Pragma" content="no-cache"><meta http-equiv="Expires" content="0">
<meta charset="utf-8">
<meta name="viewport" content="width=device-width,initial-scale=1">
-<title>FunASR 模型 — SenseVoice、Paraformer、Fun-ASR-Nano、cam++</title>
-<meta name="description" content="FunASR 模型选型指南:SenseVoice(中、粤、英、日、韩五语种)、Paraformer(中文高精度与可靠字级时间戳)、Fun-ASR-Nano(LLM 架构,vLLM 加速)、cam++(说话人分离)。">
+<title>FunASR 模型 — SenseVoice、Paraformer、Fun-ASR-Nano、MOSS、cam++</title>
+<meta name="description" content="FunASR 模型选型指南:SenseVoice(中、粤、英、日、韩五语种)、Paraformer(中文高精度与可靠字级时间戳)、Fun-ASR-Nano(LLM 架构,vLLM 加速)、MOSS(第三方转写与说话人标签)和 cam++。">
<link rel="icon" href="logo.png">
<meta property="og:title" content="FunASR 模型 — 语音识别模型对比">
-<meta property="og:description" content="FunASR 模型选型:SenseVoice、Paraformer、Fun-ASR-Nano,按速度、精度、语种选择最适合的模型。">
+<meta property="og:description" content="FunASR 模型选型:SenseVoice、Paraformer、Fun-ASR-Nano 和第三方 MOSS,按速度、精度、语种和说话人标签需求选择。">
<meta property="og:type" content="article">
<meta property="og:url" content="https://www.funasr.com/models.html">
<meta property="og:image" content="https://www.funasr.com/logo.png">
@@ -79,6 +79,7 @@ footer a{color:#60a5fa}
<tr><td><strong>Fun-ASR-MLT-Nano</strong></td><td>vLLM</td><td>31</td><td>800M</td><td>独立多语 checkpoint</td></tr>
<tr><td><strong>SenseVoice Small</strong></td><td>非自回归;需按硬件实测</td><td>中、粤、英、日、韩</td><td>234M</td><td>五语种、情感检测</td></tr>
<tr><td><strong>Paraformer-zh</strong></td><td>非自回归;需按硬件实测</td><td>中文、粤语</td><td>220M</td><td>中文最高精度</td></tr>
+<tr><td><strong>MOSS-Transcribe-Diarize</strong></td><td>批处理;NVIDIA GPU</td><td>以官方模型卡为准</td><td>第三方</td><td>长音频转写 + 说话人标签</td></tr>
<tr><td><strong>cam++</strong></td><td>实时</td><td>通用</td><td>7.2M</td><td>说话人分离与确认</td></tr>
</tbody>
</table>
@@ -162,6 +163,20 @@ result = model.generate(input="audio.wav")
print(result[0]["text"])</pre>
</div>
+<div class="model-card">
+<div class="model-name">MOSS-Transcribe-Diarize <span class="badge badge-orange">第三方模型</span></div>
+<div class="model-meta">OpenMOSS 发布 &middot; 模型 ID: <code>OpenMOSS-Team/MOSS-Transcribe-Diarize</code> &middot; <a href="https://huggingface.co/OpenMOSS-Team/MOSS-Transcribe-Diarize">HuggingFace</a> &middot; <a href="/deploy/moss-transcribe-diarize.html">FunASR 部署指南</a></div>
+<div class="model-desc">OpenMOSS 发布的第三方 Apache-2.0 模型,可在一次离线请求中生成长音频转写、时间戳和说话人标签。FunASR 提供 AutoModel、vLLM 与 SGLang Omni 的结构化结果接入;无需在服务前额外拼接 VAD,并不表示模型内部没有分段。</div>
+<div class="model-features">
+<span class="badge badge-blue">转写 + 说话人标签</span>
+<span class="badge badge-green">无需外置 VAD</span>
+<span class="badge badge-blue">vLLM / SGLang Omni</span>
+<span class="badge badge-orange">非实时</span>
+</div>
+<h3>适用场景</h3>
+<p>多人会议、访谈、播客和通话的长音频离线转写。它不是 FunASR 自有模型权重;生产前请用真实多人音频验证重叠语音、时间戳、说话人一致性和目标硬件容量。</p>
+</div>
+
<h2>辅助模型</h2>
<div class="model-card">
diff --git a/web-pages/product-site/tests/test_output.py b/web-pages/product-site/tests/test_output.py
index 3722d0e96..54b8f3fea 100644
--- a/web-pages/product-site/tests/test_output.py
+++ b/web-pages/product-site/tests/test_output.py
@@ -365,6 +365,32 @@ def test_model_pages_use_attributed_repository_routes(built_site, relative):
assert 'https://github.com/QwenAudio/SenseVoice' not in hrefs
+ ('relative', 'third_party_marker'),
+ (
+ ('models.html', '第三方模型'),
+ ('en/models.html', 'third-party model'),
+ ),
+)
+def test_model_pages_surface_moss_with_accurate_ownership_and_deployment(
+ built_site, relative, third_party_marker
+):
+ soup = read_soup(built_site / relative)
+ card = next(
+ item
+ for item in soup.select('.model-card')
+ if 'MOSS-Transcribe-Diarize' in item.get_text(' ', strip=True)
+ )
+ text = card.get_text(' ', strip=True)
+ hrefs = {link.get('href') for link in card.select('a[href]')}
+
+ assert 'OpenMOSS-Team/MOSS-Transcribe-Diarize' in text
+ assert third_party_marker in text
+ assert 'VAD' in text
+ assert 'https://huggingface.co/OpenMOSS-Team/MOSS-Transcribe-Diarize' in hrefs
+ assert ('/deploy/moss-transcribe-diarize.html' if relative == 'models.html' else '/en/deploy/moss-transcribe-diarize.html') in hrefs
+
+
@pytest.mark.parametrize('relative', ('ecosystem.html', 'en/ecosystem.html'))
def test_ecosystem_pages_use_attributed_funclip_route(built_site, relative):
soup = read_soup(built_site / relative)