diff --git a/README.md b/README.md
index 7669d7c..b72b607 100644
--- a/README.md
+++ b/README.md
@@ -3,7 +3,7 @@
Qwen3-ASR
Ready-to-use Local Speech Recognition API Service
-Speech recognition API service centered on [Qwen3-ASR](https://github.com/QwenLM/Qwen3-ASR), with NVIDIA CUDA vLLM, MetaX/MuXi MACA vLLM, and CPU Rust backends, OpenAI API compatibility, Alibaba Cloud Speech API compatibility, and a Paraformer realtime websocket capability.
+Speech recognition API service centered on [Qwen3-ASR](https://github.com/QwenLM/Qwen3-ASR), with NVIDIA CUDA vLLM, MetaX/MuXi MACA vLLM, and CPU Rust backends, OpenAI API compatibility, Alibaba Cloud Speech API compatibility, and Qwen3-ASR WebSocket streaming.
[简体中文](./docs/README_zh.md)
@@ -37,7 +37,7 @@ Speech recognition API service centered on [Qwen3-ASR](https://github.com/QwenLM
## Features
-- **Hybrid Runtime Stack** - Uses auto-selected Qwen3-ASR for offline inference and Paraformer realtime for websocket streaming
+- **Hybrid Runtime Stack** - Uses auto-selected Qwen3-ASR backends for offline inference and WebSocket streaming
- **Speaker Diarization** - Automatic multi-speaker identification using CAM++ model
- **OpenAI API Compatible** - Supports `/v1/audio/transcriptions` endpoint, works with OpenAI SDK
- **Alibaba Cloud API Compatible** - Supports Alibaba Cloud Speech RESTful API and WebSocket streaming protocol
diff --git a/docs/README_zh.md b/docs/README_zh.md
index d831140..609f1ea 100644
--- a/docs/README_zh.md
+++ b/docs/README_zh.md
@@ -3,7 +3,7 @@
Qwen3-ASR
开箱即用的本地私有化部署语音识别服务
-以 [Qwen3-ASR](https://github.com/QwenLM/Qwen3-ASR) 为核心的语音识别 API 服务，提供 NVIDIA CUDA vLLM、沐曦 MACA vLLM 与 CPU Rust 后端，兼容阿里云语音 API 和 OpenAI Audio API，并保留 Paraformer realtime WebSocket 能力。
+以 [Qwen3-ASR](https://github.com/QwenLM/Qwen3-ASR) 为核心的语音识别 API 服务，提供 NVIDIA CUDA vLLM、沐曦 MACA vLLM 与 CPU Rust 后端，兼容阿里云语音 API 和 OpenAI Audio API，并支持基于 Qwen3-ASR 的 WebSocket 流式识别。
---
@@ -35,7 +35,7 @@
## 主要特性
-- **混合运行时栈** - 离线推理由自动选择的 Qwen3-ASR 提供，WebSocket 流式由 Paraformer realtime 能力提供
+- **混合运行时栈** - 离线推理和 WebSocket 流式识别均由自动选择的 Qwen3-ASR 后端提供
- **说话人分离** - 基于 CAM++ 模型自动识别多说话人，返回说话人标记
- **OpenAI API 兼容** - 支持 `/v1/audio/transcriptions` 端点，可直接使用 OpenAI SDK
- **阿里云 API 兼容** - 支持阿里云语音识别 RESTful API 和 WebSocket 流式协议