Skip to content

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

docs: python wss server now supports multiple concurrent clients (#2985)

zhifu gaodd9487bcc376
diff --git a/runtime/quick_start.md b/runtime/quick_start.md
index 421f00fa8..c1661b926 100644
--- a/runtime/quick_start.md
+++ b/runtime/quick_start.md
@@ -11,7 +11,7 @@ You can use FunASR in the following ways:
## Service Deployment SDK
### Python version Example
-Supports real-time streaming speech recognition, uses non-streaming models for error correction, and outputs text with punctuation. Currently, only single client is supported. For multi-concurrency, please refer to the C++ version service deployment SDK below.
+Supports real-time streaming speech recognition, uses non-streaming models for error correction, and outputs text with punctuation. It now supports multiple concurrent clients with non-blocking inference (tune per-stage concurrency via --concurrent_vad / --concurrent_asr_online / --concurrent_asr_offline / --concurrent_punc / --concurrent_sv). For maximum throughput, the C++ version service deployment SDK below is still recommended.
#### Server Deployment
diff --git a/runtime/quick_start_zh.md b/runtime/quick_start_zh.md
index b472f1b0c..0317220a9 100644
--- a/runtime/quick_start_zh.md
+++ b/runtime/quick_start_zh.md
@@ -13,7 +13,7 @@
#### python版本示例
-支持实时流式语音识别,并且会用非流式模型进行纠错,输出文本带有标点。目前只支持单个client,如需多并发请参考下方c++版本服务部署SDK
+支持实时流式语音识别,并且会用非流式模型进行纠错,输出文本带有标点。现已支持多个客户端并发(非阻塞推理,可用 --concurrent_vad / --concurrent_asr_online / --concurrent_asr_offline / --concurrent_punc / --concurrent_sv 调节各阶段并发度);如需更高吞吐,仍推荐下方 c++ 版本服务部署 SDK
##### 服务端部署
```shell