full-duplex

Introduces Instruct-FD, a benchmark for evaluating whether full-duplex speech systems can follow explicit turn-taking instructions. Results show the best model achieves only 64.4% adherence, highlighting a significant gap in instruction-following turn management.

OpenAI integrates GPT-Live's full-duplex voice control into Codex and ChatGPT desktop app, enabling hands-free agentic coding with multi-threaded task execution.

This paper evaluates the reliability of Gemini models as audio judges for scoring full-duplex voice agent conversations, finding that Gemini 2.5 Flash shows strong agreement with human raters on most dimensions, though model swaps require re-validation.

OpenAI announces GPT-Live, a new full-duplex voice model that enables more natural, real-time conversations by allowing simultaneous listening and speaking, with GPT-5.5 as the backend model.

OpenAI released new full-duplex voice models GPT-Live-1 and GPT-Live-1 mini for more natural live conversations, allowing simultaneous speaking and listening, with improvements in turn-taking and context handling, and replacing Advanced Voice Mode in ChatGPT.

This paper introduces Lychee-FD, a native end-to-end full-duplex spoken language model that mitigates modality interference through a hierarchical parameter separation strategy, achieving significant improvements in speech intelligence and interaction fluidity.

TurnNat is a likelihood-based framework for automatically evaluating turn-taking naturalness in dyadic spoken dialogue, using a causal turn-taking prediction model trained on natural conversations to measure timing atypicality via negative log-likelihood.

BayLing-Duplex is a native full-duplex speech language model that enables a single autoregressive LLM to manage turn-taking and interruptions without external VAD modules, achieving high success rates and improved response quality over prior models.

This paper identifies 'state inertia' in full-duplex spoken language models, where the model's internal predictive focus lags during user interruptions, and proposes a training-free activation steering method to improve interruption handling.

Kyutai Labs released a new paper on using reinforcement learning to post-train speech models (Moshi and PersonaPlex) for more human-like interaction, including when to respond, wait, or give listening cues.

An analysis of half-duplex vs full-duplex architecture in AI voice models, discussing key features like overlap, backchannels, and barge-in that make voice agents sound robotic.

Raon-Speech is a 9B-parameter speech language model for English and Korean, supporting understanding, answering, and generation, with a full-duplex extension Raon-SpeechChat for natural real-time conversation. It achieves strong performance across 42 benchmarks and is fully open-sourced.

This paper analyzes synchronization and turn-taking dynamics in full-duplex speech dialogue models by simulating conversations between two instances of the Moshi model, measuring representational alignment via CKA and predicting turn boundaries with LSTM probes.

Thinking Machines Lab and OpenBMB released MiniCPM-o 4.5, a 9B full-duplex omnimodal model with the Omni-Flow framework that enables continuous, time-aligned real-time video and voice interaction, surpassing previous models and available as open source.

Mira Murati's team showcased a preview of the new interaction model. Trained from scratch, it natively supports full-duplex real-time audio and video conversations, instant interruptions, multi-language translation, and dynamic multi-tasking. The demonstration verified its core capabilities in low-latency streaming interaction, multimodal perception, and concurrent task execution.

EchoChain is a new benchmark for evaluating AI models' ability to revise in-progress responses when users interrupt mid-generation. The benchmark identifies three failure patterns (contextual inertia, interruption amnesia, objective displacement) and finds that across evaluated real-time voice models, no system exceeds 50% pass rate.

MoshiRAG combines a compact full-duplex speech language model with asynchronous retrieval-augmented generation to improve factuality while maintaining real-time interactivity. The approach leverages natural temporal gaps in conversation to retrieve external knowledge without disrupting the natural flow of dialogue.

MTR-DuplexBench introduces a comprehensive benchmark for evaluating Full-Duplex Speech Language Models in multi-round conversations, addressing challenges like blurred turn boundaries and context inconsistency while assessing conversational features, dialogue quality, instruction following, and safety.

OpenAI has released the new generation ChatGPT voice feature powered by GPT-Live, enabling full-duplex real-time conversations, supporting interruption, intelligent reasoning, web search, and real-time translation, making interactions more natural and intelligent.