Skip to content

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

docs: clarify native vLLM transcription path

LauraGPTa9c40696375e
diff --git a/docs/vllm_guide_zh_v2.md b/docs/vllm_guide_zh_v2.md
index f05646b87..ad0c76208 100644
--- a/docs/vllm_guide_zh_v2.md
+++ b/docs/vllm_guide_zh_v2.md
@@ -84,12 +84,26 @@ model = AutoModelVLLM(
#### B. vLLM 原生转写路径
-vLLM 的支持模型表把
+vLLM 原生 `FunASRForConditionalGeneration` 路径使用完整的 native checkpoint,
+不是上文的 FunASR `model.pt` 布局。官方维护的模型为
+[`FunAudioLLM/Fun-ASR-Nano-2512-vllm`](https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512-vllm):
+
+```bash
+vllm serve FunAudioLLM/Fun-ASR-Nano-2512-vllm --port 8000
+```
+
+vLLM 的支持模型表也曾使用
[`allendou/Fun-ASR-Nano-2512-vllm`](https://huggingface.co/allendou/Fun-ASR-Nano-2512-vllm)
-列为原生 `FunASRForConditionalGeneration` 架构的示例。这是托管在官方
-FunAudioLLM 组织之外的社区转换完整 checkpoint。只有明确选择 vLLM 原生转写
-接口时,才应按该模型卡与 vLLM 文档使用它;不要用它替换下文 FunASR
-`AutoModelVLLM` 示例中的官方 checkpoint。
+这一社区转换示例。它同样必须通过 `vllm serve` 的原生接口使用;不要把任一
+native checkpoint 传给 `AutoModelVLLM` 或
+`examples/industrial_data_pretraining/fun_asr_nano/serve_realtime_ws.py`。后两者预期
+官方 `model.pt`、`config.yaml` 与 `Qwen3-0.6B/` 目录,因而会拒绝 native layout。
+
+原生 `vllm serve` 的转写 API 不会自动获得 FunASR WebSocket 服务的 VAD、partial
+预览、会话状态或说话人处理。需要实时行为时,请按 vLLM 的
+[realtime speech-to-text 示例](https://docs.vllm.ai/en/latest/examples/speech_to_text/realtime/)
+在应用侧实现音频分段、请求调度与可替换预览;需要 FunASR 已实现的 WebSocket
+协议时,继续使用上文的官方 checkpoint + `serve_realtime_ws.py` 路径。
**硬件**:GPU ≥ 8GB VRAM,CUDA ≥ 11.8。推荐 16GB+。
diff --git a/tests/test_vllm_model_source_docs.py b/tests/test_vllm_model_source_docs.py
index e77edd1de..14c43c3ff 100644
--- a/tests/test_vllm_model_source_docs.py
+++ b/tests/test_vllm_model_source_docs.py
@@ -29,3 +29,17 @@ def test_vllm_guides_distinguish_official_and_native_model_paths(relpath):
for marker in required_markers:
assert marker in text, f"{relpath} is missing {marker}"
+
+
+def test_v2_guide_keeps_native_vllm_and_funasr_realtime_paths_separate():
+ text = (ROOT / "docs/vllm_guide_zh_v2.md").read_text(encoding="utf-8")
+
+ required_markers = [
+ "FunAudioLLM/Fun-ASR-Nano-2512-vllm",
+ "vllm serve FunAudioLLM/Fun-ASR-Nano-2512-vllm --port 8000",
+ "serve_realtime_ws.py",
+ "realtime speech-to-text",
+ ]
+
+ for marker in required_markers:
+ assert marker in text, f"docs/vllm_guide_zh_v2.md is missing {marker}"