From 006d1d785642e01f358f105970e24bb0d063e1ec Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E6=B8=B8=E9=9B=81?= Date: Sun, 24 May 2026 19:39:41 +0800 Subject: [PATCH] docs: add vLLM engine and dynamic VAD to READMEs - README.md: add vLLM/dynamic VAD to What's new, add vLLM row to features - README_zh.md: same updates in Chinese - Fun-ASR-Nano README/README_zh: add built-in vLLM and WebSocket service links --- README.md | 3 +++ README_zh.md | 2 ++ examples/industrial_data_pretraining/fun_asr_nano/README.md | 2 ++ examples/industrial_data_pretraining/fun_asr_nano/README_zh.md | 2 ++ 4 files changed, 9 insertions(+) diff --git a/README.md b/README.md index dfbb303fb..996ea6e3d 100644 --- a/README.md +++ b/README.md @@ -61,6 +61,7 @@ That's it. **One model, one call** — VAD segmentation, speech recognition, pun | Emotion | ✅ Happy/Sad/Angry | ❌ | ❌ | | Languages | 50+ | 57 | Varies | | Streaming | ✅ WebSocket | ❌ | ✅ | +| vLLM Acceleration | ✅ 2-3x faster | ❌ | N/A | | Self-hosted | ✅ MIT license | ✅ MIT license | ❌ Cloud only | | Cost | Free | Free | $0.006/min+ | | CPU viable | ✅ 17x realtime | ❌ Too slow | N/A | @@ -87,6 +88,8 @@ That's it. **One model, one call** — VAD segmentation, speech recognition, pun ## What's new +- 2026/05/24: **vLLM Inference Engine** — 2-3x faster LLM decoding for Fun-ASR-Nano. Streaming WebSocket service with VAD + Speaker Diarization. [Guide →](docs/vllm_guide.md) +- 2026/05/24: **Dynamic VAD** — adaptive silence threshold (default on). Short sentences stay intact, long segments get auto-split. [Details →](docs/vllm_guide.md#附录dynamicstreamingvad) - 2026/05/24: **v1.3.3** — `funasr-server` CLI, OpenAI-compatible API, MCP Server for AI agents. `pip install --upgrade funasr` - 2026/05/20: Added Qwen3-ASR (0.6B/1.7B) — 52 languages, auto detection. [usage](examples/industrial_data_pretraining/qwen3_asr) - 2026/05/20: Added GLM-ASR-Nano (1.5B) — 17 languages, dialect support. [usage](examples/industrial_data_pretraining/glm_asr) diff --git a/README_zh.md b/README_zh.md index 15e9fe6b3..ca06813a2 100644 --- a/README_zh.md +++ b/README_zh.md @@ -87,6 +87,8 @@ result = model.generate(input="meeting.wav") ## 最新动态 +- 2026/05/24：**vLLM 推理引擎** — Fun-ASR-Nano 解码加速 2-3 倍。支持流式 WebSocket 服务（VAD + 说话人分离 + 热词）。[文档 →](docs/vllm_guide.md) +- 2026/05/24：**动态 VAD** — 自适应静音阈值（默认开启），短句不切碎、长句自动切分。[详情 →](docs/vllm_guide.md#附录dynamicstreamingvad) - 2026/05/24：**v1.3.3** — `funasr-server` 命令行工具、OpenAI 兼容 API、MCP 服务。`pip install --upgrade funasr` - 2026/05/20：新增 Qwen3-ASR (0.6B/1.7B)，52 种语言自动检测。[使用方法](examples/industrial_data_pretraining/qwen3_asr) - 2026/05/20：新增 GLM-ASR-Nano (1.5B)，17 种语言，方言优化。[使用方法](examples/industrial_data_pretraining/glm_asr) diff --git a/examples/industrial_data_pretraining/fun_asr_nano/README.md b/examples/industrial_data_pretraining/fun_asr_nano/README.md index 08c7a61ab..8c3e83763 100644 --- a/examples/industrial_data_pretraining/fun_asr_nano/README.md +++ b/examples/industrial_data_pretraining/fun_asr_nano/README.md @@ -197,6 +197,8 @@ We evaluated Fun-ASR against other state-of-the-art models on open-source benchm ## Remarkable Third-Party Work +- **Built-in vLLM Inference Engine**: 2-3x faster decoding with AutoModelVLLM, streaming WebSocket service, tensor parallel. [Guide →](../../../docs/vllm_guide.md) +- **Streaming WebSocket Service**: Real-time ASR with VAD + Speaker Diarization + Hotwords. [Quick Start →](docs/realtime_demo.md) - vLLM (GPU) Deployment Best Practices: An accelerated implementation of Fun-ASR using vLLM. [Repository](https://github.com/yuekaizhang/Fun-ASR-vllm) ## Citations diff --git a/examples/industrial_data_pretraining/fun_asr_nano/README_zh.md b/examples/industrial_data_pretraining/fun_asr_nano/README_zh.md index 0757f305d..2c18530e9 100644 --- a/examples/industrial_data_pretraining/fun_asr_nano/README_zh.md +++ b/examples/industrial_data_pretraining/fun_asr_nano/README_zh.md @@ -196,6 +196,8 @@ if __name__ == "__main__": ## 优秀三方工作 +- **内置 vLLM 推理引擎**：AutoModelVLLM 解码加速 2-3 倍，流式 WebSocket 服务，tensor parallel 多卡并行。[文档 →](../../../docs/vllm_guide.md) +- **流式 WebSocket 服务**：实时语音识别 + VAD 分句 + 说话人分离 + 热词定制。[快速上手 →](docs/realtime_demo.md) - vLLM (GPU) 最佳部署实践: 使用 vLLM 实现对 Fun-ASR 的加速. [Repository](https://github.com/yuekaizhang/Fun-ASR-vllm) ## Citations