turn-detection 开源项目推荐（45 个）

TEN-framework /ten-turn-detection

用于全双工对话通信的轮次检测

alexkroman /tiny-audio

一个只需 25 美元即可训练的语音转文本系统。冻结的语音编码器与 Qwen LLM 由一个训练好的小型投影器连接，支持词级时间戳、说话人分离、批处理 HTTP 服务器、实时演示和免费课程。

videosdk-live /NAMO-Turn-Detector-v1

面向对话式 AI 的高性能语义轮次检测

wananing /voice-agent-handbook

实时语音 agent 架构手册。四种链路形态（级联 / 半级联 / S2S / 全双工）的取舍；判停、打断、首音、工具回合、会话恢复、音频前处理、评测七个机制的设计空间；pipecat、livekit-agents、TEN、unmute、moshi、Qwen3-Omni、CosyVoice、FunASR、xiaozhi 等 28 个开源项目在这些机制上的实现，分析到文件和函数并标注 commit。附框架与模型对比矩阵、选型决策树。

sshah03 /CueLight

一种始终监听的语音 agent 轮次检测器，小到可在每 20 毫秒音频上运行。

justplus /turn-detection

人机对话轮次检测模型，有效解决声学VAD等待过长的问题

wavekat /wavekat-turn

Turn detection library for Rust with a unified trait interface over multiple backends (Pipecat Smart Turn, LiveKit EOU). Part of the WaveKat voice pipeline.

xcc-zach /xturnix

Streaming dialogue turn prediction model based on textual chat history

dangvansam /livekit-plugins-turn-detector

Turn detector plugin for LiveKit Agents with external model (Triton, OpenAI)

dangvansam /livekit-plugins-namo-turn-detector

Turn detector plugin for LiveKit Agents with Namo model

Crele-AI /nimbus

Crele Nimbus: a real-time audio decision model. Ask yes/no questions about live audio in plain English.

miuda-ai /smart-turn-rs

A Rust library for livekit's smart turn detection v3 using ONNX Runtime.

kadirnar /voice-agent-next

Real-time speech-to-speech voice agents: native S2S models and streaming cascades, local and cloud, on Linux, macOS and Windows — with a built-in benchmark suite.

santhosh-005 /smart-turn-livekit

LiveKit Agents plugin for Smart Turn end-of-turn detection

saolabsai /cue

Streaming CPU turn detection for voice agents. Rust, Python and Pipecat with Smart Turn and Silero VAD.

rockbenben /kids-ai

面向儿童的 Android 原生语音聊天应用：孩子没说完绝不抢话，完全离线优先保隐私，支持自带大模型密钥直连

deepgram-starters /rust-flux

Get started using Deepgram's Flux API for real-time transcription with turn detection using this Rust demo app

bubustack /livekit-turn-detector-engram

Turn detector Engram for bobrapet — buffers PCM audio, runs WebRTC VAD, and emits speech turn payloads.

deepgram-starters /fastapi-flux

Get started using Deepgram's Flux real-time transcription with this FastAPI demo app

deepgram-starters /bun-flux

Get started using Deepgram's Flux real-time transcription with this Bun demo app