---
title: "Google's Gemini 3.5 Live Translate Brings Real-Time Voice Translation to 70+ Languages"
subtitle: "Google opens Gemini 3.5 Live Translate to developers via the Live API, enabling continuous speech-to-speech translation across 70+ languages with just a few lines of config"
url: "https://alphasignal.ai/news/google-s-gemini-3-5-live-translate-brings-real-time-voice-translation-to-70"
author: "AlphaSignal Newsroom"
author_url: "https://alphasignal.ai/editorial-team"
company: "Google AI Developers"
company_url: "https://alphasignal.ai/company/googleaidevs"
published: "2026-06-12T21:36:42.000Z"
updated: "2026-06-16T19:19:03.354Z"
type: "News"
topics:
  - "Audio"
  - "Api"
subtopics:
  - "Realtime Voice"
  - "Speech To Text"
  - "Text To Speech"
upvotes: 262
read_time: "6 min"
source_url: "https://accounts.google.com/v3/signin/identifier?continue=https%3A%2F%2Faistudio.google.com%2Flive%3Fmodel%3Dgemini-3.5-live-translate-preview&dsh=S588575593%3A1781019809431122&followup=https%3A%2F%2Faistudio.google.com%2Flive%3Fmodel%3Dgemini-3.5-live-translate-preview&passive=1209600&flowName=WebLiteSignIn&flowEntry=ServiceLogin&ifkv=AcDsRvxApVmpp-QCkTsm_btKFoa0OLsZq8DTIP5h-tlFyxm2Cp3ove4Ys4dpoFbbgyxIpEw_S6BJGQ"
comments: 0
---

# Google's Gemini 3.5 Live Translate Brings Real-Time Voice Translation to 70+ Languages

> Google opens Gemini 3.5 Live Translate to developers via the Live API, enabling continuous speech-to-speech translation across 70+ languages with just a few lines of config

Real-time voice translation has been a demo-stage promise for years. You'd see it on stage, marvel at it, and then go back to subtitles. Google is now making a serious push to change that: [Gemini 3.5 Live Translate](https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-live-3-5-translate/) is a dedicated speech-to-speech model that any developer can wire into their app today via the [Gemini Live API](https://ai.google.dev/gemini-api/docs/live-api/live-translate).

## One model, three surfaces

Google launched Gemini 3.5 Live Translate as a streaming speech-to-speech model that translates across 70+ languages with automatic detection.
It is rolling out simultaneously across three distinct channels:

- Developers get public preview access through the Gemini Live API and Google AI Studio.
- Enterprises get private preview access in Google Meet starting in June.
  Google Meet leaps from 5 English-only languages to 2,000+ combinations.
- The consumer app update is live globally on Android and iOS, with no sign-up required.

Google shipped a headphone-based beta on Android in December 2025 in the United States, Mexico, and India, expanded to iOS in March 2026, and has now delivered the remaining pieces: developer access through the Gemini Live API and Google AI Studio, plus an enterprise integration in Google Meet.
The developer API is the new and most significant unlock here.

## Under the hood: one model, not a pipeline

The architecture is the real story.
Unlike traditional voice AI stacks that chain separate speech-to-text, language model, and text-to-speech components, the Live API processes audio end-to-end in a single model -- audio in, audio out.
That matters enormously for latency. Every hop in a classic pipeline adds delay and a new failure mode. Collapsing it into one model cuts both.

The model is built on Gemini 3 Pro and streams speech-to-speech translation in over 70 languages, auto-detecting the source language and generating output that preserves the speaker's intonation, pacing, and pitch.
According to the Gemini 3.5 Audio model card, it accepts audio input with up to a 128K-token context window and produces audio and text output up to 64K tokens.

Turn-by-turn systems wait for a speaker to finish before responding. Gemini 3.5 Live Translate generates speech continuously instead. It balances a trade-off between waiting for context and translating immediately -- more context improves quality, while faster output keeps the translation in sync with the speaker. The result stays a few seconds behind the speaker throughout a session.

The model is also not a general-purpose assistant. Google's docs are explicit about this distinction:

| Live Agent | Live Translate |
| --- | --- |
| Acts as an assistant; reasons and takes actions | Acts as an interpreter; pure translation pipeline |
| Turn-based; handles interruptions and intent | Continuous stream processing; no turn detection |
| Supports tools, function calling, system instructions | No tools, no instructions -- translation only |
| Fully multimodal (text, audio, video, image) | Audio input only |

## Wiring it up takes minutes

Developers configure it via the Live API using `targetLanguageCode` and `echoTargetLanguage`; audio-only, 16kHz in, 24kHz out.
The full session setup in Python looks like this:

```
from google import genai
from google.genai import types
client = genai.Client()
model = "gemini-3.5-live-translate-preview"
config = types.LiveConnectConfig(
    response_modalities=["AUDIO"],
    input_audio_transcription=types.AudioTranscriptionConfig(),
    output_audio_transcription=types.AudioTranscriptionConfig(),
    translation_config=types.TranslationConfig(
        target_language_code="es",  # translate to Spanish
        echo_target_language=True   # echo back if speaker is already in Spanish
    )
)
async with client.aio.live.connect(model=model, config=config) as session:
    # stream raw PCM audio chunks at 16kHz
    # receive translated audio at 24kHz + synced transcripts
    ...
```

The `echoTargetLanguage` flag is a subtle but useful detail: when set to `true`, the model will pass through audio that is already in the target language rather than going silent. For two-way conversations where both speakers might use the target language at times, this keeps the output stream consistent. You can also enable synchronized text transcripts for both input and output simultaneously, which is what the demo in the original tweet showcases with live radio broadcast ingestion.

## Who is already building with it

Developer platforms like Agora, Fishjam, LiveKit, Pipecat, and Vision Agents have integrated the Gemini Live API, enabling developers to build and deploy voice translation apps with ease. These integrations handle the complex real-time media streaming infrastructure, so developers can focus on the user experience.

Grab is testing the model to enable multilingual communication in near real-time between drivers and travelers at pickups. These users make over 10 million voice calls per month through Grab.
That is a meaningful stress test at scale.

## Where it struggles

Google's own [model card](https://deepmind.google/models/model-cards/gemini-3-5-audio/) is refreshingly candid about the limitations. Do not treat this as a solved problem:

- **Voice consistency:** Voices can shift after long pauses, change gender, or get stuck on one voice during rapid multi-speaker sessions.
- **Language detection edge cases:** The model struggles with heavy accents, similar languages (Spanish vs. Portuguese), or rapid language switches.
- **Background audio:**
  Google says the model handles background noise and overlapping voices, making it usable in real-world settings like busy cafes or open offices
  -- but not all background audio is filtered cleanly.
- **No text input:** Only audio input is supported. You cannot feed it text and get translated speech out.
- **No tool use:** Unlike the general Live API, Live Translate does not support function calling or system instructions.

## Safety and compliance baked in

All generated audio carries an imperceptible SynthID watermark for detectability.
The EU AI Act's Article 50, requiring synthetic content labeling, comes into force in August 2026. SynthID watermarking puts Google ahead of that deadline.
For anyone building production apps in regulated markets, that is not a footnote -- it is a compliance checkbox already ticked.

## What this actually unlocks

The new step is the combination: near real-time speech-to-speech translation across more than 70 languages, with automatic language detection, continuous output, tone preservation, and product distribution through Google Translate, Google Meet, and the Gemini Live API.
That bundle is what makes this different from prior attempts.

The practical use cases that are now genuinely viable:

- **Multilingual customer support lines** -- route calls without knowing the caller's language in advance
- **Live broadcast and radio ingestion** -- the exact demo Google showed, translating a foreign-language broadcast in real time
- **International voice agents** -- build a voice bot that handles 70+ languages without separate per-language models
- **Cross-language meetings** -- embed translation directly into video call infrastructure
- **Accessibility tools** --
  a new listening mode on Android streams translated audio through the device earpiece, allowing users to privately follow translations without headphones

It is a preview release, so Google has not committed to production SLAs, stable pricing, or general availability yet.
But the API is open, the model ID is `gemini-3.5-live-translate-preview`, and you can start prototyping in [Google AI Studio](https://aistudio.google.com/live?model=gemini-3.5-live-translate-preview) right now without writing a line of code. For a capability that has been a science fiction staple for decades, that is a meaningful moment.

## Sources

- [X post by @googleaidevs](https://x.com/googleaidevs/status/2065548876753654106)
- [Live translation with Gemini Live API | Google AI for Developers](https://ai.google.dev/gemini-api/docs/live-api/live-translate)
- [Gemini 3.5 Audio (Live Translate) - Model Card — Google DeepMind](https://deepmind.google/models/model-cards/gemini-3-5-audio/)
- [Fluid, natural voice translation with Gemini 3.5 Live Translate — Google Blog](https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-live-3-5-translate/)
- [Google Expands AI Live Speech Translation with Gemini 3.5 Live Translate — Slator](https://slator.com/google-ai-live-speech-translation-gemini-3-5-live-translate/)
- [Google Releases Gemini 3.5 Live Translate — MarkTechPost](https://www.marktechpost.com/2026/06/09/google-releases-gemini-3-5-live-translate-a-streaming-speech-to-speech-audio-model-covering-70-languages-across-meet-translate-and-the-live-api/)