GPT-Live-1 vs Gemini 3.8 TTS: uno sostiene una conversazione, l'altro legge un copione — e solo uno dei due è economico

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.

Prima, il problema del nome, perché decide il prezzo che quoti.

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 — il livello completo, oltre 130 lingue, 0,50 $ per milione di token di testo in input e 9,00 $ per milione di token audio in output, che saliranno a 1,00 $ e 18,00 $ il 1° gennaio 2027.

• gemini-3.8-flash-lite-tts — 101 lingue, stessa tariffa di input, 6,00 $ per milione di token audio in uscita, che salgono a 12,00 $.

• Entrambi prevedono tariffe Batch e Flex a metà del prezzo Standard ($0,25 in input, $4,50 o $3,00 in output) e tariffe Priority a 1,8× ($0,90 in input, $16,20 o $10,80 in output).

L'audio viene fatturato a 25 token al secondo, ed è da qui che deriva il valore al minuto: 25 token × 60 secondi = 1.500 token audio al minuto, quindi 9,00 $ per milione di token corrispondono a circa 0,0135 $ al minuto. Flash-Lite si attesta vicino a 0,009 $.

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.

Cosa fa effettivamente ciascun modello quando lo chiami

Gemini 3.8 Flash TTS prende in input del testo e restituisce audio. Non ha alcun input audio, nessun turno da prendere, nessuna nozione di essere interrotto e nessuna trascrizione da restituire, perché non ha mai sentito nulla. L'endpoint unario restituisce WAV a 24 kHz mono; lo streaming restituisce PCM a 16 bit senza intestazione, con mulaw e alaw accettati e una frequenza di campionamento configurabile. Le indicazioni sceniche inline vengono scritte nello script come eventi tra parentesi angolari — <laugh>, <sigh>, <breath>, <short pause> — e il testo di input viene per il resto trattato come una trascrizione letterale, non come istruzioni. L'API Live non supporta affatto questi modelli.

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.

I numeri, e di chi sono

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 1.275 al rango 5, con un intervallo del 95% all'incirca tra 1.260 e 1.290 su 2.783 campioni, indicato a 16,5 $ per milione di caratteri.

• Gemini 3.8 Flash-Lite TTS — Elo 1.248 al rango 8, intervallo circa 1.234 a 1.262, 11,0 $ per milione di caratteri.

• 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.

Il design vocale è il punto in cui i due convergono e dove tornano a divergere.

Entrambi i fornitori ora ti permettono di descrivere una voce a parole. È una novità del 2026 ed è il punto in cui i due prodotti si avvicinano di più a sovrapporsi.

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.

Due differenze decidono quale puoi effettivamente usare:

• 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.

Dove il divario di prezzo smette di essere 27×

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.

• La lunghezza dell'output è controllabile da un lato. Un lavoro di rendering è una quantità nota: puoi preventivare un articolo di 900 parole prima di generarlo, e le tariffe Batch lo dimezzano ulteriormente. Una conversazione no.

• 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.

Quale scegliere

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.

Se stai sostenendo una conversazione, il confronto TTS è un errore di categoria e la domanda onesta è diversa: il backend delegato produce una risposta abbastanza veloce da mantenere breve la sessione? Questo è il numero che decide la tua fattura, ed è quello che nessuno mette su una slide di lancio.

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.

Niente di tutto ciò è ipotetico: i fallimenti documentati dell'IA e gli incidenti degli agenti alla base di rilasci come questo sono raccolti in l'OrcaRouter AI Incident Archive, dove ogni voce riporta le proprie fonti, la gravità, il grado di confidenza e il flag di contestazione.