GPT-Live-1 語音現在可用描述的：Prompt-to-Voice 正式推出，九款國際預設語音同步登場

For most of the year, choosing a voice for G​PT-Live-1 meant picking a string. You read a list, you set audio.output.voice/ to one of them at session start, and if none of them fit the brand you were building, you shipped with the closest one. That changed on 16 October 2026, when O​penAI released prompt-to-voice to all API customers and added nine natural voices aimed at five non-English markets. The new capability is small on paper — one endpoint, free to call — and it rewrites the earliest decision most voice products make. It also lands a day after G​PT-Live-1.1 picked up native Decisions integration inside the same Live sessions, which matters because the two changes push in opposite directions: one makes the voice more bespoke, the other makes the thing behind the voice more active.

以下是實際推出的內容、其成本，以及新語音清單中那兩個沒有任何發布說明提及的不對稱之處。

10月16日發行版包含的內容

兩件事，都標記到 gpt-live-1/ 模型以及兩個端點——v1/audio/voices/ 與 v1/live/sessions/。

• Prompt-to-voice — create a custom voice from a text description, through the API or the voice Playground, then reuse it in G​PT-Live sessions. Released for all API customers.

• 九款全新自然語音 — Nuri、Juni 和 Flitz（德語）；Brise 和 Sillage（法語）；Nira（印地語）；Harema 和 Shida（日語）；Noeul（韓語）。

• Cost — creating a prompt-generated voice is free. Using it speech-synthesises at regular G​PT-Live session rates, and auditioning it in the Playground counts as a normal session for billing.

• A day earlier, on 15 October — native Decisions integration in G​PT-Live, so an app can act while the user is still speaking or react to what Live says, via native delegation or transcript deltas.

The pricing context that makes those numbers legible: a G​PT-Live-1 voice session runs $0.05 per minute, billed by the second and not rounded up. That line is unchanged by this release, and no part of prompt-to-voice changes it.

API 呼叫就是一次 POST，而它的形式會告訴你模型正在做什麼

你將 JSON 傳送至 /v1/audio/voices，並附上 type: "prompt"、一個 name、prompt的描述，以及 "model": "auto"。可選的 script_hint最多 2,000 個字元，提供範例語音來引導生成；若省略，服務會自行撰寫腳本，但空白或過短的腳本會被拒絕。

回應中不包含音訊。它只回傳中繼資料——一個物件類型、一個 id/、name/、一個以 Unix 秒為單位的 created_at/，type/，以及你最初輸入的提示詞。這是刻意的設計選擇，並帶來一個實際後果：建立時沒有東西可以預覽，所以你只能像實際使用時那樣試聽，而這會以一次工作階段計費。每個請求都會儲存一個全新的語音，因此請把 ID 和描述放在一起——描述是唯一記錄了你當初要求什麼的地方。

The controllable qualities are all expressed as ordinary language rather than as separate parameters: accent, age, presentation, pitch, timbre, texture, loudness, speech rate and distinctiveness. O​penAI's own examples include an accent phrased as "American English, with a strong accent" and a pace phrased as "Medium pace." There is no numeric prosody slider anywhere in the surface, which means the prompt is the API.

那確實是實質的取捨。滑桿可重現；一句話則不行。兩個團隊寫出同一段描述，未必會得到相同的聲音，而文件也直言，生成的聲音不保證獨一無二。請把提示文字納入版本控制，就像你管理系統提示一樣。

值得在設計時納入考量的限制

• 每個組織 100 個由提示生成的語音。此上限也適用於您的客戶，因此如果您要為每位終端使用者生成一個語音，100 是您整個租用戶基礎的上限，而不是每個帳戶的配額。

• G​PT-Live only. Prompt-generated voices do not work with the Text to Speech, Realtime, or Chat Completions APIs. A voice you describe today is a voice you cannot reuse in a batch narration pipeline tomorrow.

• 無法與 Zero Data Retention 搭配使用。透過 /v1/audio/voices 建立語音（包括錄音路徑）均不適用。

• 僅限成人，且不得模仿。不得使用兒童的聲音，指引原則是描述特質，而非指名真實人物或角色。

• 以錄音為基礎的語音是另一條更嚴格的途徑——每個組織最多 20 個，樣本長度為 30 秒或以下，格式可為 mpeg、wav、ogg、aac、flac、webm 或 mp4，且至少要有五秒的真實語音與十五個轉錄詞元，上傳檔案上限為 10 MiB，並須提供同意錄音。

The 100-voice cap and the G​PT-Live-only restriction are the two that shape architecture. If you are a platform generating a voice per customer, you have a hard ceiling and no fallback path into the cheaper speech APIs.

完整的預設清單，以及它不足的地方

The nine additions land on top of a library that O​penAI's Live voice guide now lists twenty-one deep, plus marin/ as the default. Read as a whole, the list says something about who O​penAI expects to be building:

• 英語——六種地區色彩，而非一種。Quartz 和 Ripple 屬於澳洲；Vesper 屬於英國；Willow 和 Stone 屬於愛爾蘭；Gleam 和 Meridian 屬於北美；Beacon 屬於菲律賓；Delta 和 Cinder 屬於美國南部。其中四個標記為「生成」而非「自然」。

• 葡萄牙語 — Bossa 與 Tempo，兩者皆為巴西腔，兩者皆為 Natural。

• 德語 —— Nuri、Juni 與 Flitz。兩個陰性，一個陽性。

• 法式 — Brise 與 Sillage。一款女性，一款男性。

• 日文 — Harema 和 Shida。一個陰柔，一個陽剛。

• 印地語 — Nira。僅限陰性。

• 韓文 — Noeul。僅限女性。

Two asymmetries fall out of that. First, German gets three voices while French, Japanese, Hindi and Korean get two, two, one and one — an allocation that tracks O​penAI's European market priorities more than it tracks speaker populations. Second, Hindi and Korean each get exactly one voice, and both are feminine. If your product needs a masculine Hindi or Korean voice from the preset library, there isn't one, and your options are a recording-based voice or a prompt-generated one. That is precisely the gap prompt-to-voice was released to fill, arriving the same day — which suggests the sequencing is not a coincidence.

指南也指出了一件容易誤解的事：「區域影響」描述的是某個語音的說話風格，而不是它能說哪些語言的硬性限制。請用你的應用程式實際遇到的語言和發音來測試候選語音，而不是假設標示為德語的語音就是德語輸入的正確選擇。

通話中無法變更的設定。 語音是啟動設定。你指派 audio.output.voice/ 當你建立工作階段時，若要變更它，就得啟動新的工作階段。分支會繼承來源工作階段的模型、原始指令與輸入，因此比較兩種語音意味著從相同設定建立兩個工作階段，而不是在同一個工作階段內切換。如果你要對真實使用者進行語音 A/B 測試，那就代表需要兩個工作階段工廠，而不是一個帶有旗標的工作階段——而且這代表每次比較都會在兩個測試組上耗費完整的工作階段時間。

後端部署在哪裡，以及為什麼那是這個決策中較便宜的一半

The voice is the half of a Live session that users hear. The other half is delegation: G​PT-Live listens and speaks while a backend model or agent does the reasoning and tool work, and backend usage is billed separately at normal rates. O​penAI's own cost guidance makes the arithmetic blunt — full active session time is billable including while the backend works, and using a faster backend model with parallel tool calls so the user finishes a minute earlier saves $0.05 in voice charges. The voice rate is fixed per minute; the backend is the dial.

That is where a router earns its place rather than being bolted on. Client delegation lets you point a G​PT-Live session at your own agent, harness or service — which means the delegated backend can be any model, and the useful version of "any model" is one key and one endpoint rather than a contract per provider. OrcaRouter fronts 200-plus models at provider list price with no markup, so a backend price cut reaches you the same day it lands, and automatic failover covers a provider having a bad afternoon without the user hearing it. The routing DSL lets you compose the reasoning step out of several models instead of betting the conversation on one, which is the shape you want for a delegated backend that has to answer inside a live conversation's latency budget.

One boundary to state plainly: OrcaRouter does not host G​PT-Live-1, and nothing in this release changes that. The router's role in a Live architecture is behind the delegation boundary, not in front of the voice.

本週該做什麼

如果你已經用預設語音推出 Live 產品，這九項新增功能就是免費的機會，讓你測試非英語使用者是否真的獲得服務——在一小部分工作階段中更換語音並傾聽，但請記住，每次更換都會讓每個實驗組耗掉一個新的工作階段。

如果你先前卡在品牌契合度上，現在值得花一天實驗 prompt-to-voice，但有一個之後會反咬你一口的注意事項：把描述文字當成建置產物。對它做版本控制，把回傳的 ID 保存在它旁邊，並假設你將會需要重新產生——因為在組織上限為 100 個語音、且建立時無法預覽的情況下，你週二喜歡的那個語音，是你必須能夠刻意重現的東西。

And if you are building the whole system rather than just the voice, the more consequential October change may be the one nobody wrote a blog post about. A day before the voices landed, G​PT-Live picked up native Decisions integration — the ability to act while the user is still mid-sentence. A voice that fits is a product decision. A backend that answers before the user finishes asking is an architecture one.

這些都不是假設：像這次發布這類版本背後那些有紀錄的 AI 失效與代理事件，全都收錄在OrcaRouter AI Incident Archive，其中每筆條目都附有自己的來源、嚴重程度、信心等級與爭議標記。