turn-detection

Turn detection for full-duplex dialogue communication

A speech-to-text system you can train for $25. A frozen speech encoder and a Qwen LLM joined by a small trained projector, with word timestamps, speaker diarization, a batched HTTP server, a live demo, and a free course.

High-performance, semantic turn detection for conversational AI

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。

An always-listening turn detector for voice agents, small enough to run on every 20 ms of audio.

人机对话轮次检测模型，有效解决声学VAD等待过长的问题

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

Streaming dialogue turn prediction model based on textual chat history

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

Turn detector plugin for LiveKit Agents with Namo model

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

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

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.

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

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

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

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

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

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

Demonstrate using Deepgram's Flux with a Bun app to build and run speech recognition projects quickly and efficiently.