GPT-Live-1 vs Gemini 3.8 TTS: 하나는 대화를 이어가고, 하나는 대본을 읽는다 — 그리고 둘 중 하나만 저렴하다

Set G​PT-Live-1 and Gemini 3.8 Flash TTS next to each other on a per-minute meter and the answer looks obvious. G​oogle's narrator bills about 1.35 cents per minute of generated speech at the 2026 rate. O​penAI's conversational model bills 5 cents per minute of open line — and keeps billing while the backend thinks, while the user pauses, and while nothing is being said at all. Roughly a twenty-seven-fold gap, before anyone has asked whether the two are doing the same job.

They are not, and in a way that is more interesting than the usual "different categories" hand-wave. Gemini 3.8 Flash TTS cannot hear. G​PT-Live-1 cannot be batched. And the naming makes this worse than it needs to be, because the phrase G​emini 3.8 TTS — the label half the internet is using — does not correspond to any model you can call.

먼저 이름 문제부터 봅시다. 제시하시는 가격을 좌우하니까요.

There is no model called G​emini 3.8 TTS. G​oogle's own documentation uses the phrase as umbrella language for two callable endpoints, and they are priced differently enough that quoting the wrong one is a real error:

• gemini-3.8-flash-tts — 풀 티어, 130개 이상 언어, 텍스트 입력 100만 토큰당 $0.50, 오디오 출력 100만 토큰당 $9.00이며, 2027년 1월 1일에는 $1.00과 $18.00으로 인상됩니다.

• gemini-3.8-flash-lite-tts — 101개 언어, 동일한 입력 요금, 오디오 출력 토큰 100만 개당 $6.00, $12.00까지 상승.

• 둘 다 Batch 및 Flex 요금을 Standard 가격의 절반($0.25 입력, $4.50 또는 $3.00 출력)으로 제공하며, Priority 요금은 1.8배($0.90 입력, $16.20 또는 $10.80 출력)입니다.

오디오는 초당 25토큰으로 청구되며, 분당 수치가 여기서 나옵니다: 25토큰 × 60초 = 분당 1,500 오디오 토큰이므로, 백만 개당 $9.00는 분당 약 $0.0135가 됩니다. Flash-Lite는 약 $0.009에 가깝습니다.

The status of the model is genuinely unsettled and worth saying out loud, because G​oogle's own surfaces disagree. The G​emini API changelog calls both models generally available. The API model list labels them "New Stable." G​oogle Cloud's text-to-speech documentation describes them as available in Preview through the G​emini Enterprise API only. The announcement blog post, published on 23 September 2026, uses neither word. Anyone writing a procurement note this month should treat the GA label as the changelog's claim rather than a settled fact.

각 모델을 호출할 때 실제로 수행하는 작업

Gemini 3.8 Flash TTS는 텍스트를 입력받아 오디오를 반환합니다. 오디오 입력도, 주고받을 차례도, 중단된다는 개념도, 돌려줄 전사본도 없습니다. 애초에 아무것도 듣지 못하기 때문입니다. 단일(unary) 엔드포인트는 24kHz 모노 WAV를 반환하고, 스트리밍은 헤더 없는 16비트 PCM을 반환하며, mulaw와 alaw도 지원되고 샘플 레이트는 설정할 수 있습니다. 인라인 무대 지시는 스크립트에 꺾쇠괄호 이벤트로 작성됩니다 — <laugh>, <sigh>, <breath>, <short pause> — 그 외의 입력 텍스트는 지시문이 아니라 그대로의 전사본으로 취급됩니다. Live API는 이러한 모델을 전혀 지원하지 않습니다.

G​PT-Live-1 does the opposite job. It is full duplex: it processes incoming audio while producing outgoing audio, decides several times a second whether the user has finished talking, and returns speech-recognition transcripts as part of the session. Its inputs are audio and text; its outputs are audio and text; images and video are unsupported. It streams and supports function calling but not structured outputs, and knowledge cut-off is 31 July 2025. Only one endpoint serves it — v1/live/sessions — and Chat Completions, Responses, Realtime and Batch are all out.

That last detail is the one that ends the comparison for a lot of workloads. If your product renders a script, G​PT-Live-1 is neither the cheap option nor the correct one. If your product answers a phone, Gemini 3.8 Flash TTS is one third of a pipeline that still needs an ASR model and a language model bolted on either side — and the whole point of G​PT-Live-1 is that the seam between those three parts is where conversations break.

숫자들, 그리고 그것들이 누구의 것인지

On Artificial Analysis's Provider Voice Arena — a blind-vote Elo board over each vendor's own native voices — G​oogle's new models land mid-field rather than at the top:

• Gemini 3.8 Flash TTS — Elo 1,275로 5위, 2,783개 표본에 걸쳐 약 1,260~1,290의 95% 구간, 백만 자당 $16.5로 책정.

• Gemini 3.8 Flash-Lite TTS — Elo 1,248, 순위 8위, 구간은 대략 1,234~1,262, 백만 글자당 $11.0.

• Above both — Eleven v4 Turbo at 1,330, Eleven v4 at 1,329, Q​wen-Audio-3.1-TTS-Plus at 1,296 and Cartesia Sonic 3.6 at 1,285.

G​oogle's own claim is stronger than that board, and it is a vendor claim: the 23 September post says the models took first place on Hume AI's Voice Design Benchmark at 71.4 and led Hume's accent modelling at 60.8, plus top blind-preference positions in Japanese, Brazilian Portuguese, Vietnamese, Modern Standard Arabic, Mexican Spanish and Hindi. Those are G​oogle-reported placements on someone else's evaluation, not a reproduction.

Flipping to the conversational side, G​PT-Live-1's numbers come from a different board entirely. Artificial Analysis's speech-to-speech index — a weighted average of speech reasoning, agentic performance, arena preference and task success — now lists G​PT-Live-1 under named backend configurations rather than as one row: G​PT-Live-1 (Astra, medium) and G​PT-Live-1 (Sol, low) sit at the top of that index, with G​rok Voice 3.0 leading the field. That is a useful reminder that in this architecture the voice model is only half the row; the delegated backend moves the score.

One number that matters and does not exist anywhere: G​oogle publishes no official latency figure for either TTS model. Every time-to-first-audio number in circulation is third-party — one aggregator's p50 sits around 3.47 seconds for Flash TTS and 2.66 for Flash-Lite — and none of it is directly comparable to G​PT-Live-1's conversational cadence, because they measure different things.

보이스 디자인은 이 둘이 수렴하는 지점이자, 다시 발산하는 지점이다

이제 두 업체 모두 목소리를 말로 설명할 수 있게 해줍니다. 이는 2026년에 새로 생긴 기능이며, 두 제품이 서로 겹치는 부분에 가장 가까이 다가간 지점입니다.

G​oogle's path runs through the G​emini API: a prompt-based voice persona saved persistently as a voice_ ID, voice replication from about a 30-second sample with a consent check, SynthID watermarking and C2PA credentials attached, and a cap of 200 stateful custom voices per project with one-year retention from last use. The written description covers role, accent and vocal traits.

O​penAI's path, released on 16 October 2026 to all API customers, is a POST to /v1/audio/voices with type: "prompt", a name, the description and "model": "auto". Creation is free; using the voice bills at normal session rates. It returns metadata only — no audio, no preview. The qualities you can steer are accent, age, presentation, pitch, timbre, texture, loudness, speech rate and distinctiveness, all expressed as ordinary language with no numeric parameters.

두 가지 차이점이 실제로 어떤 것을 사용할 수 있는지를 결정합니다:

• Prompt-generated voices on G​PT-Live work only with G​PT-Live. They are explicitly incompatible with the Text to Speech, Realtime and Chat Completions APIs. G​oogle's custom voices are, by contrast, voices for a rendering engine, which is what makes them reusable across a catalogue.

• The organisation cap on O​penAI's side is 100 prompt-generated voices, and it applies to voices generated on behalf of your customers too. G​oogle's ceiling is 200 per project. If you are a platform minting a voice per customer, both numbers are your real product limit, not a footnote.

There is also a difference in what you get back. A G​oogle voice arrives with SynthID and C2PA provenance attached. An O​penAI prompt-generated voice arrives as an ID, a creation timestamp and your own prompt text — and the documentation says outright that generated voices are not guaranteed to be unique. Keep the description under version control; it is the only record of what you asked for.

가격 차이가 27배가 아니게 되는 지점

The per-minute comparison is the one that gets quoted and it is the one that misleads, because G​PT-Live-1's meter is not measuring the same resource. Three things move the real ratio:

• Idle time is billable. A voice session is billed per second for as long as it is open, including while the backend model is working and while the user is silent. G​oogle's meter only advances when audio is generated. A support call where the user talks for four minutes and your backend thinks for two is six billable minutes on Live and fewer on the renderer — before you have counted the ASR and language-model legs the renderer still needs.

• 출력 길이는 한쪽에서는 제어할 수 있습니다. 렌더링 작업은 예측 가능한 항목입니다. 900단어짜리 글은 생성하기 전에 가격을 책정할 수 있고, Batch 요금을 적용하면 비용이 다시 절반으로 줄어듭니다. 대화는 그렇지 않습니다.

• The backend is a separate bill. Delegated reasoning and tool use on G​PT-Live-1 are charged at normal model rates on top of the $0.05 a minute, and O​penAI's own cost guidance frames the lever as making the user finish sooner — for example, running a faster backend with parallel tool calls so the conversation ends a minute earlier, saving $0.05. The voice rate is fixed; the session length is the dial.

That last point is where a routing layer stops being a nice-to-have. Client delegation lets you point a G​PT-Live session at your own agent or service, which means the delegated backend can be any model — and the useful version of "any model" is one endpoint and one key rather than a contract per provider. OrcaRouter fronts 200-plus models at provider list price with no markup, so a vendor price cut on the backend side reaches you the day it lands, and automatic failover covers a provider outage without the caller hearing it. If you are building the delegated path, the backend choice is where your cost actually lives; the $0.05 a minute is a constant.

To be precise about the boundary: OrcaRouter does not host G​PT-Live-1 or the G​emini 3.8 TTS models. What it routes is the reasoning side of a delegated Live session.

어느 것을 선택할까

If you are rendering — articles, documentation, video narration, IVR prompts written in advance — Gemini 3.8 Flash TTS is the correct tool and G​PT-Live-1 is roughly 27 times too expensive per audio minute and structurally incapable of the job in batch. Take Flash-Lite if 101 languages cover you, take Batch or Flex rates if latency is not the constraint, and model the January 2027 price doubling from the start, because G​oogle has published it in the same table rather than burying it.

대화를 나누고 있다면, TTS 비교는 범주 오류이고, 정직한 질문은 다릅니다: 위임된 백엔드가 세션이 짧게 유지될 만큼 충분히 빠르게 답을 내놓는가? 그 수치가 당신의 청구서를 결정하며, 아무도 출시 슬라이드에 넣지 않는 바로 그 수치입니다.

And if you are simply trying to work out what "G​emini 3.8 TTS" costs — the answer is $9.00 per million audio tokens, on one of the two models that name refers to, until the end of 2026.

그중 어느 것도 가정이 아닙니다: 이번과 같은 릴리스의 배경이 된 문서화된 AI 실패와 에이전트 사고는 OrcaRouter AI Incident Archive에 수집되어 있으며, 각 항목에는 자체 출처, 심각도, 신뢰도 등급, 분쟁 플래그가 함께 제공됩니다.