Gemini 3.5 Transcribe Live

by Google · Released , 13 days after Gemini 3.7 Flash · Announcement · All Google releases

Gemini 3.5 Transcribe Live is the streaming variant of Gemini 3.5 Transcribe, Google's speech-to-text suite released on August 26, 2026. It runs on the Gemini Live API as gemini-3.5-transcribe-live: clients stream raw 16-bit, 16kHz mono PCM audio in roughly 100ms chunks and receive speculative interim hypotheses while someone speaks, followed by a finalized transcript when the turn ends. Output is text only. Turn handling supports automatic server-side voice activity detection, a hybrid mode that finalizes faster using client-side silence detection, and manual push-to-talk signaling. Google says the model detects language automatically across more than 85 languages, handles live language switching, supports custom vocabulary biasing of up to 1,000 terms, and offers a smart transcription mode that removes filler words and formats text. On Google's numbers, Gemini 3.5 Transcribe averages 4.0% word error rate in streaming (2.6% non-streaming) and is 70% faster than Chirp 3. Live sessions are capped at 10 minutes and do not include speaker diarization or word-level timestamps, which the non-streaming model offers. It fits voice agents, live captioning, dictation, and voice-driven interfaces that need text as the user talks.

Sign up without a credit card · Realtime WebSocket
Text input
$3.50 / 1M
Text output
$21.00 / 1M
Audio output
$21.00 / 1M
Routes
2 EU · US

Available routes

Gemini 3.5 Transcribe Live runs on 2 different routes through the Opper gateway.

ProviderRegionZero data retentionTrainingInputOutputCache read
USNo$3.50$21.00Input price
EUNo$3.50$21.00Input price

Compare residency, zero data retention and training posture at a glance, with full data-handling detail per route below. In the EU it is served by Google Cloud. See every Gemini model hosted in the EU.

Live routes and prices as JSON: api.opper.ai/v3/models?q=gemini%2Fgemini-3.5-transcribe-live

Key info

Input
Output
Features
Text input price
$3.50 /1M
Text output price
$21.00 /1M
Audio input price
$3.50 /1M
Audio output price
$21.00 /1M
Released
  • EU residency available on Google Cloud
  • US residency available on Google Cloud
  • Zero data retention with abuse monitoring
  • No training by default

Uptime and availability

Gemini 3.5 Transcribe Live runs on a single route through the Opper gateway today, so its availability is that provider's availability.

99.95%route uptime, last 30 days
Single route, no failover target within this model.

Name a second model in the same request and the gateway tries it on retriable errors, so a busy hour never has to reach your users. Set up a fallback chain.

Measured over the last 30 days from each provider's official status feed via StatusGator. Refreshed hourly. See uptime for every provider Opper monitors.

Data handling per route

Each route hosting Gemini 3.5 Transcribe Live has its own privacy posture, residency, and GDPR terms. Postures are maintained by Opper with a last-verification timestamp.

Google Cloud — United States🇺🇸

Zero data retention with abuse monitoring: prompts and outputs are kept for abuse monitoring only, for a window the provider does not state, and are not used for training. No training on customer data. US; SCCs.

Zero data retention
With abuse monitoring, window not stated.
Training
No training on customer data.
Logging
Abuse monitoring
Abuse monitoring
On by default, holds flagged content, 55 days, a person can view samples
Transfer mechanism
SCCs

Google Cloud — EU🇪🇺

Zero data retention with abuse monitoring: nothing is logged or used for training; content flagged by abuse monitoring is held for review. No training on customer data. EU.

Zero data retention
With abuse monitoring, flagged content held.
Training
No training on customer data.
Logging
None
Abuse monitoring
On by default, holds flagged content, 90 days, a person can view samples
Transfer mechanism
Not applicable — data stays in EU

Get started

Call Gemini 3.5 Transcribe Live through the Opper gateway with one API key. Realtime models run over a WebSocket at wss://api.opper.ai/v3/realtime: open it, send session.start with this model, then stream audio both ways.

See the Opper realtime quickstart, the realtime protocol and the Opper multimodal skill.

Gemini 3.5 Transcribe Live is a premium model on Opper. Sign up needs no credit card: you get an API key straight away and the free models work in the playground and the API. Add a card to use premium models, pay-as-you-go with no minimum.

Set it up with your agent

Copy this and paste it into a coding agent like Claude Code, Cursor or Codex and it'll wire up Opper for you.

Or call it directly

import WebSocket from "ws";
const ws = new WebSocket("wss://api.opper.ai/v3/realtime", {
headers: { Authorization: `Bearer ${process.env.OPPER_API_KEY}` },
});
ws.on("open", () => {
ws.send(JSON.stringify({
type: "session.start",
config: {
model: "gemini/gemini-3.5-transcribe-live",
},
}));
});
// Stream microphone audio in with { type: "audio.append", audio: <base64 PCM16> }
ws.on("message", (raw) => {
const ev = JSON.parse(raw.toString());
if (ev.type === "transcript.committed") console.log(ev);
});

Use Gemini 3.5 Transcribe Live in your apps

Pick your tool, then choose Gemini 3.5 Transcribe Live to see how to run it through Opper.

Compare Gemini 3.5 Transcribe Live with…

Side-by-side on privacy, EU hosting, pricing, and benchmarks.

Other models from Google

Start building with 700+ models

One API key for every major provider, up and running in minutes.

Get startedView Documentation