Subvolts take Google's new model translates spoken conversations live with natural voices. Sep 16, 2026 This is a new model release.
- Developers
- Integrate real-time voice translation into applications for global audiences.
- Business
- Facilitate smoother international communication and collaboration with clients and partners.
This model enables real-time, natural-sounding voice translation, improving cross-lingual communication for users.
translate Gemini 3.5 Live Translate
Spec sheet arrives with the next pipeline run. The source page has the details.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
Google's Gemini 3.5 Live Translate model offers low-latency, real-time audio-to-audio translation for voice conversations. It provides seamless, bidirectional translation with high accuracy and natural-sounding voice output. This preview aims to improve cross-lingual communication.
Imagine talking to someone who speaks a different language, and an AI instantly translates your words back and forth. Google's new Gemini 3.5 Live Translate does just that, making conversations smoother and more natural.
This model targets real-time voice translation, a challenging area for AI. It could significantly impact global business communication and personal interactions. Competitors like OpenAI and Meta are also investing in voice and translation technologies. Future developments will focus on expanding language support and reducing latency further for truly seamless conversations.
- Model Name: Gemini 3.5 Live Translate
- Company: Google
- Type: Low-latency audio-to-audio translation
- Functionality: Real-time voice conversation translation
- Output: Natural voice output
Google's Gemini 3.5 Live Translate enters the real-time voice translation space, competing with ongoing advancements from OpenAI and Meta in voice AI.
- low-latency
- Refers to a system with minimal delay between input and output.
- Responds very quickly with almost no waiting time.
- audio-to-audio model
- A machine learning model that takes audio input and produces audio output, often involving processing and transformation.
- A program that listens to sounds and then creates new sounds.
The Playbook: reusable how-tos, prompts, configs and tips pulled out of the news. From the Playbook, matched to the product and topics of this item.
- ToolGemini API Live for real-time voice and videoAPI for low-latency real-time voice and video interactions with Gemini 2.5 Flash. · 15 min
- ToolGemini 3.1 Flash TTS for Low Latency SpeechGenerates natural-sounding speech with low latency and expressive control using audio tags. · 5 min
- Model pickGemini 3.8 Flash-Lite TTS for low latency text to speechUse Gemini 3.8 Flash-Lite TTS for everyday tasks needing fast, natural-sounding speech synthesis. · 5 min
- Model pickChoose Gemini 3.8 Flash-Lite TTS for production workloadsSelect Gemini 3.8 Flash-Lite TTS for high-throughput production needs, replacing older preview models. · 5 min
Questions people ask
- What is Gemini 3.5 Live Translate?
- It is a low-latency audio-to-audio model from Google optimized for real-time voice conversation translation.
- What are the key features?
- It offers seamless bidirectional translation with high accuracy and natural voice output.
More from Gemini Models 471 more
Page generated Oct 19, 2026. Summaries are Subvolts' own; the story belongs to Gemini Models.