full-duplex

Plug-and-play streaming semantic VAD for real-time full-duplex spoken dialogue systems.

A curated list of full-duplex spoken dialogue models & benchmarks

Jarvis × Codex v0.2.0：macOS 透明全息语音助手，本机“嗨 Jarvis”唤醒，直连 Codex Voice，在同一线程中对话、执行任务并支持 STOP 中断。

[EMNLP 2026 Main] Speaking While Listening: a survey and empirical audit of full-duplex spoken dialogue systems — L0–L3 architectural hierarchy, T×I×R interaction ontology, and a five-state decision machine, with a curated list of models, datasets, and benchmarks.

PersonaPlex on Apple Silicon: an MLX port of NVIDIA’s full-duplex speech-to-speech model with realtime local/web modes and offline WAV inference.

Peer-to-peer, secure, TCP/UDP port forwarding using HTTP(s) relay for NAT/firewall traversal

Self-hosted realtime voice gateway for Hermes Agent. Keep talking while Hermes runs background tasks through local, Gemini Live, and OpenAI Realtime adapters.

Hermes Talk: realtime duplex voice for Hermes Agent. Rides your ChatGPT or X Premium subscription, no API key needed. OpenAI, Grok, Gemini. Terminal, Discord, dashboard.

Adaptive filters for GNU Radio

Arbos: an open-source, file-system-native agent coordinator with a Cursor-style desktop, full-duplex voice, iPhone app, and a mesh of workers

WebView attachable full-duplex asynchronous interoperable independent messaging library between .NET and JavaScript.

Linux kernel multi- bidirectional- logical- channel communication driver for connecting User/Kernel space entities via full-duplex symmetrical transport.

DSH 全双工语音插件：流式语音识别入草稿、语音合成按句朗读 + 实时字幕、开口即打断；本地识别免 API Key，可选唤醒词。 · Full-duplex voice plugin for DeepSeek Harness: streaming speech-to-text into an editable draft, text-to-speech read-aloud with live captions, true barge-in and wake word; on-device ASR, no API key.

Game-Time: Evaluating Temporal Dynamics in Full-Duplex Spoken Language Models

Real-time, full-duplex AI voice bot integrating NVIDIA's PersonaPlex with Twilio Media Streams for natural speech-to-speech conversations.

Simplest and fully-customizable RPC standalone infrastructure.

Real-time serving of full-duplex interaction models: per-tick deadline scheduling + KV-budget admission control. Systems research.

Set of 📝 with 🔗 to help those building realtime (S2S) voice ai 🎙️🤖

600ms-budget full-duplex voice agent — cascade ASR·RAG·LLM·TTS (458ms) + end-to-end Omni (250ms) | Voice · ASR · TTS · LLM · Infra · Posttraining

Real-time streaming talking-head avatar: PCM audio in, MuseTalk lip-synced video out over a self-developed WebSocket transport. Full-duplex voice sessions with pluggable ASR / LLM / TTS spokes and ms-level barge-in.