Controllable Transcription. Verbatim ( every, filler, pause, stutter, vocal sound) , or intended ( what the speaker meant to say, optimized for readability) with word-level timestamps.
-
Updated
Nov 11, 2026 - Python
Controllable Transcription. Verbatim ( every, filler, pause, stutter, vocal sound) , or intended ( what the speaker meant to say, optimized for readability) with word-level timestamps.
Turn detection for full-duplex dialogue communication
Real-time Turn-taking, Backchannel, and Head-nodding Prediction for Full-duplex Interaction
DuplexJev-32B and DuplexJev-4B: speech decision models for full-duplex voice agents. Typed single-token readout from a frozen speech LLM, served with vLLM. Try it: api.adventists.cn/duplexjev
[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.
System-One-controlled full-duplex voice interaction: turn-taking as plain-language decisions judged by Jev.
End-to-end spoken-dialogue framework: deterministic semantic speech units, unified interleaved token streams with turn-control tokens, a duplex turn-taking state machine with barge-in and latency budgets, deterministic session replay, and turn-taking evaluation on planted synthetic dialogues.
Crele Nimbus: a real-time audio decision model. Ask yes/no questions about live audio in plain English.
全双工语音助手:边听边说、打断、任务委派与工具执行。Full-duplex voice agent with barge-in, streaming audio, and async thinking. DuplexOmni (arXiv:2606.09186).
Local-first VAD, barge-in, and turn-taking primitives for interruptible voice agents.
Official code for DuplexPO (NeurIPS 2026): RL for conversational dynamics in full-duplex spoken dialogue models - dynamics-critical windows, the FCDR reward, and event-level GRPO, framework-neutral
Python package based on Gibson's framework (2003) for turn-taking in group conversation analysis.
EgoSpeak: Learning When to Speak for Egocentric Conversational Agents in the Wild
Full-duplex conversational turn-taking in C++17 — no ASR, no LLM, no neural VAD. Decides when you have finished speaking from prosody alone, and hears you interrupt it through its own speaker echo.
🏀💬 Advanced User Interfaces course project developed as a 1st year student of MSc in CS Eng. at Politecnico di Milano
A voice agent whose interaction model is visible and swappable: turn-taking, barge-in, and a background model that reports back on its own. ASR, TTS, foreground and background each run on any OpenAI-compatible endpoint.
Cue: when a voice agent should stop talking, and when it should answer. Barge-in and end-of-turn decisions from the caller's audio.
Dhairya (धैर्य): a turn-taking layer on Gnani Prisma v3.0 that lets a voice agent wait while a caller finishes reading out a number.
Full-duplex spoken dialogue research: fine-tuning a speech model (LFM3-Audio-2B, QLoRA) toward an offline voice assistant that listens while it talks, on a 4 GB laptop GPU. Paused; code, frozen test definitions, results and a review paper.
Production C/C++ TurnSense runtime using canonical GGUF and upstream ggml
To associate your repository with the turn-taking topic, visit your repo's landing page and select "manage topics."