Build more natural voice experiences with GPT‑Live‑1 in the API

通过 API 构建更自然的语音体验：GPT‑Live‑1 正式发布

GPT‑Live‑1 brings ChatGPT’s natural, full-duplex conversations to the API, with more control over how voice agents speak and act. GPT‑Live‑1 将 ChatGPT 自然的全双工对话能力引入 API，让开发者能够更精准地控制语音代理的言谈与行为。

We’re launching GPT‑Live‑1 in the API, giving developers a powerful, natural voice model for building voice-enabled apps and business workflows. First introduced in ChatGPT, GPT‑Live‑1 is capable of listening and speaking at the same time, and, as seen with Codex and ChatGPT Work, can delegate deeper reasoning and actions to the models and tools it is paired with. 我们正式在 API 中推出 GPT‑Live‑1，为开发者提供了一个强大且自然的语音模型，用于构建语音应用和业务工作流。GPT‑Live‑1 最初在 ChatGPT 中亮相，它能够同时进行听取和说话。正如在 Codex 和 ChatGPT Work 中所展示的那样，它还可以将更深层的推理和操作任务委派给与之配合的模型和工具。

For the API release of GPT‑Live‑1, we’ve focused on new capabilities that let developers steer and customize voice experiences around their users, workflows, and goals. A core GPT‑Live‑1 strength, smooth interruption handling, is already delivering business impact: in early evaluations, Speak found that GPT‑Live‑1 gave learners more time to think before the language tutor responded, cutting interruptions by almost 80% versus previous turn-based systems. 在 GPT‑Live‑1 的 API 版本中，我们重点推出了多项新功能，让开发者能够围绕用户、工作流和目标来引导并定制语音体验。GPT‑Live‑1 的核心优势之一是流畅的中断处理，这已经产生了实际的业务影响：在早期评估中，Speak 发现 GPT‑Live‑1 为学习者留出了更多的思考时间，在语言导师做出回应前，将中断次数比之前的轮次制系统减少了近 80%。

Key strengths of GPT‑Live‑1 in the API:

GPT‑Live‑1 API 的核心优势：

Interruption handling: Improves interruption handling via a single model that reasons over incoming and outgoing audio together, avoiding the latency and brittle handoffs of chained STT–LLM–TTS architectures. 中断处理： 通过单一模型同时处理输入和输出音频，从而改进了中断处理能力，避免了传统 STT–LLM–TTS 链式架构中常见的延迟和脆弱的交接问题。
Reasoning & tool calling delegation: GPT‑Live‑1 can delegate reasoning and tool calls to a backend text model like GPT‑6 Astra or a third-party model. 推理与工具调用委派： GPT‑Live‑1 可以将推理和工具调用任务委派给后端文本模型（如 GPT‑6 Astra）或第三方模型。
Tone, pace, and style: Lets developers shape an agent’s tone, pace, and conversational style through the system prompt. 语调、语速与风格： 允许开发者通过系统提示词（System Prompt）来塑造代理的语调、语速和对话风格。
Silent context management & background noise: Better handles background noise and silence without interrupting the conversation or narrating every step out loud. 静默上下文管理与背景噪音： 能够更好地处理背景噪音和静默，而不会中断对话或大声叙述每一个步骤。
Long-session reliability: Improves context retention and conversational quality across extended interactions. 长会话稳定性： 提升了在长时间交互中的上下文留存能力和对话质量。
Telephony support: Enables deployment of full-duplex voice agents for phone calls, from restaurant reservations to customer support. 电话支持： 支持部署用于电话通话的全双工语音代理，涵盖从餐厅预订到客户支持的各类场景。

Simplify your voice-agent architecture and reduce voice latency

简化语音代理架构并降低语音延迟

Traditional voice agents stitch together speech-to-text, a reasoning model, and text-to-speech. Each handoff adds latency and creates more opportunities to lose timing, context, or the natural rhythm of a conversation. Developers are often the ones left coordinating those stages, including what happens when someone interrupts, pauses, or changes direction. 传统的语音代理通过拼接语音转文字（STT）、推理模型和文字转语音（TTS）来实现。每一次交接都会增加延迟，并增加了丢失时机、上下文或对话自然节奏的风险。开发者往往需要亲自协调这些阶段，包括处理用户中断、停顿或改变话题时的情况。

GPT‑Live‑1 handles listening and speaking in a single model, simplifying the voice layer. It can respond to interruptions and acknowledgements as they happen, while delegating deeper reasoning to the back end. This lets the conversation continue while work happens in the background. GPT‑Live‑1 在单一模型中处理听取和说话，简化了语音层。它可以在中断和确认发生时立即做出响应，同时将更深层的推理任务委派给后端。这使得对话可以在后台处理任务的同时继续进行。

“Compared to our cascaded build, GPT‑Live‑1 simplified our code base by 80% and removed 23K lines of code. This enabled natural, real-time patient conversations & freed our team to improve the experience from booking an appointment to navigating care.”—Tony Stoyanov, Co-Founder & CTO “与我们之前的级联架构相比，GPT‑Live‑1 将我们的代码库简化了 80%，并移除了 2.3 万行代码。这实现了自然、实时的患者对话，并让我们的团队能够专注于改善从预约挂号到就医引导的全流程体验。”——Tony Stoyanov，联合创始人兼 CTO

Developers choose the models, tools, and agent harness behind the conversation. For example, they might pair GPT‑Live‑1 with a model like Luna for high-volume tasks like scheduling or order updates, and use a model like Astra for complex customer issues that require reasoning. That flexibility lets developers match reasoning depth, speed, and cost to each task. 开发者可以自主选择对话背后的模型、工具和代理框架。例如，他们可以将 GPT‑Live‑1 与 Luna 等模型搭配用于调度或订单更新等高频任务，并使用 Astra 等模型处理需要复杂推理的客户问题。这种灵活性使开发者能够根据每项任务的需求，匹配相应的推理深度、速度和成本。

GPT‑Live‑1 natively provides ASR transcripts and response text. It also offers strong alphanumeric understanding and supports keyword biasing. Although GPT‑Live‑1 is not a turn-based model, it natively supports turn detection, so developers can continue to build around explicit turn boundaries. GPT‑Live‑1 原生提供 ASR 转录文本和响应文本。它还具备强大的字母数字理解能力，并支持关键词偏好设置。尽管 GPT‑Live‑1 不是轮次制模型，但它原生支持轮次检测，因此开发者仍然可以基于明确的轮次边界进行构建。

Measuring the full-duplex advantage

衡量全双工优势

Across our evaluations, GPT‑Live‑1 improves Full Duplex Bench performance by 30 percentage points over GPT‑Realtime‑2.1, with large gains in turn-taking latency and interactive behavior. Paired with GPT‑6 Astra at medium reasoning effort, it also ranks #1 on Tau3, which measures frontier voice-agent intelligence on end-to-end tasks. 在我们的评估中，GPT‑Live‑1 的 Full Duplex Bench 性能比 GPT‑Realtime‑2.1 提升了 30 个百分点，在轮次切换延迟和交互行为方面取得了显著进步。当与 GPT‑6 Astra（中等推理强度）配合使用时，它在衡量前沿语音代理端到端任务智能的 Tau3 测试中排名第一。