Google DeepMind Blog

Gemini 3.1 Flash Live: Making audio AI more natural and reliable

3.5内容质量
Gemini 3.1 Flash Live: Making audio AI more natural and reliable

TL;DR · AI 摘要

本文仅抓取了Google DeepMind博客的导航与元数据,正文内容缺失。标题虽提及Gemini 3.1 Flash Live致力于提升音频AI的自然度与可靠性,但全文无架构解析、性能基准或工程落地指南,信息密度极低,不具备工程师阅读价值。

核心要点

  • 提供的文本仅为网页导航与元数据,缺失模型核心正文。
  • 标题表明该版本聚焦于优化音频AI的交互自然度与系统可靠性。
  • 无技术原理、评测数据或API调用示例,无法指导实际工程开发。
#Gemini#音频AI#Google DeepMind#大语言模型
打开原文

Gemini 3.1 Flash Live: Google’s latest AI audio model

Skip to main content

The Keyword

Gemini 3.1 Flash Live: Making audio AI more natural and reliable

Share

x.comFacebookLinkedIn[Mail](mailto:?subject=Gemini%203.1%20Flash%20Live%3A%20Making%20audio%20AI%20more%20natural%20and%20reliable&body=Check%20out%20this%20article%20on%20the%20Keyword:%0A%0AGemini%203.1%20Flash%20Live%3A%20Making%20audio%20AI%20more%20natural%20and%20reliable%0A%0AGemini%203.1%20Flash%20Live%20is%20now%20available%20across%20Google%20products.%0A%0Ahttps://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-1-flash-live/)

Copy link

Innovation & AI

Learn more:

See all AI updates

[See all](http://deepmind.google/innovation-and-ai/models-and-research/ "See all Models & Research articles")

[See all](http://deepmind.google/innovation-and-ai/products/ "See all Products articles")

[See all](http://deepmind.google/innovation-and-ai/infrastructure-and-cloud/ "See all Infrastructure & cloud articles")

Learn more:

Google DeepMind blogGoogle Research blogGoogle Developers blogGoogle Cloud blog

See all AI updates

  • Products & platforms

Products & platforms

Learn more:

See all product updates

[See all](http://deepmind.google/products-and-platforms/products/ "See all Products articles")

[See all](http://deepmind.google/products-and-platforms/platforms/ "See all Platforms articles")

[See all](http://deepmind.google/products-and-platforms/devices/ "See all Devices articles")

Learn more:

Google Ads & Commerce blogWaze blog

See all product updates

  • Company news

Company news

[See all](http://deepmind.google/company-news/outreach-and-initiatives/ "See all Outreach & initiatives articles")

[See all](http://deepmind.google/authors/ "See all Leadership articles")

[See all](http://deepmind.google/company-news/inside-google/ "See all Inside Google articles")

Subscribe

["How is Gemini changing Maps?", "What is \"vibe design?\"", "How can I learn new AI skills?"]

Search freely using keywords, or by asking a question

Suggested searches

Subscribe

The Keyword

Innovation & AI

Learn more:

See all AI updates

  • Products & platforms

Products & platforms

Learn more:

See all product updates

  • Company news

Company news

  • [Images](http://deepmind.google/image-library/ "Images")
  • [RSS feed](http://deepmind.google/rss/ "RSS feed")

Subscribe

Breadcrumb

  1. [](https://blog.google/ "The Keyword")
  2. Innovation & AI
  3. Models & research
  4. Gemini Models

Gemini 3.1 Flash Live: Making audio AI more natural and reliable

Mar 26, 2026

· 9 min read

Share

x.comFacebookLinkedIn[Mail](mailto:?subject=Gemini%203.1%20Flash%20Live%3A%20Making%20audio%20AI%20more%20natural%20and%20reliable&body=Check%20out%20this%20article%20on%20the%20Keyword:%0A%0AGemini%203.1%20Flash%20Live%3A%20Making%20audio%20AI%20more%20natural%20and%20reliable%0A%0AGemini%203.1%20Flash%20Live%20is%20now%20available%20across%20Google%20products.%0A%0Ahttps://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-1-flash-live/)

Copy link

Our latest voice model has improved precision and lower latency to make voice interactions more fluid, natural and precise.

Valeria Wu

Product Manager

Yifan Ding

Software Engineer on behalf of the Gemini team

Read AI-generated summary

General summary

Gemini 3.1 Flash Live is Google's highest-quality audio model, designed for natural and reliable real-time dialogue. Developers can access it through the Gemini Live API in Google AI Studio, while enterprises can use it for customer experience. Everyone can experience it via Search Live and Gemini Live, which now supports over 200 countries.

Summaries were generated by Google AI. Generative AI is experimental.

Bullet points

  • "Gemini 3.1 Flash Live" is here, making AI audio sound more natural and reliable.
  • This new audio model is faster and better at understanding tone for natural conversations.
  • Developers can use it to build voice agents that handle complex tasks more reliably.
  • Gemini Live and Search Live now offer more helpful responses in many languages.
  • All audio from 3.1 Flash Live is watermarked to help prevent the spread of misinformation.

Summaries were generated by Google AI. Generative AI is experimental.

#### Explore other styles:

  • General summary
  • Bullet points

Share

x.comFacebookLinkedIn[Mail](mailto:?subject=Gemini%203.1%20Flash%20Live%3A%20Making%20audio%20AI%20more%20natural%20and%20reliable&body=Check%20out%20this%20article%20on%20the%20Keyword:%0A%0AGemini%203.1%20Flash%20Live%3A%20Making%20audio%20AI%20more%20natural%20and%20reliable%0A%0AGemini%203.1%20Flash%20Live%20is%20now%20available%20across%20Google%20products.%0A%0Ahttps://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-1-flash-live/)

Copy link

Image 2: The Gemini emblem sits next to text reading 'Gemini 3.1 Flash Live'. The background has blue, multicolored dots making up a microphone icon
Image 2: The Gemini emblem sits next to text reading 'Gemini 3.1 Flash Live'. The background has blue, multicolored dots making up a microphone icon

Your browser does not support the audio element.

Listen to article

This content is generated by Google AI. Generative AI is experimental

[[duration]] minutes

Voice Speed

Voice

Speed 0.75X 1X 1.5X 2X

Today, we’re advancing Gemini’s real-time dialogue capabilities with Gemini 3.1 Flash Live, our highest-quality audio and voice model yet. It delivers the speed and natural rhythm needed for the next generation of voice-first AI, offering a more intuitive experience for developers, enterprises and everyday users.

3.1 Flash Live is available across Google products:

For developers: Robust reasoning and task execution

We’ve improved 3.1 Flash Live’s overall quality, making it more reliable for developers and enterprises to build voice-first agents that can complete complex tasks at scale. On ComplexFuncBench Audio, a benchmark that captures multi-step function calling with various constraints, it leads with a score of 90.8% compared to our previous model.

Image 3: ComplexFuncBench audio bar graph
Image 3: ComplexFuncBench audio bar graph
Image 4: BigBenchAudio bar graph
Image 4: BigBenchAudio bar graph

On Scale AI’s Audio MultiChallenge, Gemini 3.1 Flash Live leads with a score of 36.1% with “thinking” on. The benchmark specifically tests complex instruction following and long-horizon reasoning amidst the interruptions and hesitations typical of real-world audio.

Image 5: AudioMultiChallenge bar graph
Image 5: AudioMultiChallenge bar graph

3.1 Flash Live also has improved tonal understanding to deliver more natural dialogue. In Gemini Enterprise for Customer Experience, it’s even more effective at recognizing acoustic nuances like pitch and pace than 2.5 Flash Native Audio. It’s also better at dynamically adjusting its response to users' expressions of frustration or confusion.

Sorry, your browser doesn't support embedded videos, but don't worry, you can download it and watch it with your favorite video player!

Read more

3.1 Flash Live lets you build voice-ready agents that handle complex tasks in noisy environments.

Illustrative demonstration built with Gemini 3.1 Pro, powered by Gemini 3.1 Flash Live.

Sorry, your browser doesn't support embedded videos, but don't worry, you can download it and watch it with your favorite video player!

Read more

3.1 Flash Live lets you use your voice to vibe code and quickly iterate.

Illustrative demonstration built with Gemini 3.1 Pro, powered by Gemini 3.1 Flash Live.

Jump to position 1 Jump to position 2

Companies like Verizon, LiveKit and The Home Depot have given positive feedback on 3.1 Flash Live in their workflows, highlighting its improved, natural conversation.

Image 6: Quote from The Home Depot
Image 6: Quote from The Home Depot
Image 7: Quote from Verizon
Image 7: Quote from Verizon
Image 8: Quote from LiveKit
Image 8: Quote from LiveKit
Image 9: Quote from Wavera
Image 9: Quote from Wavera
Image 10: Quote from Stream
Image 10: Quote from Stream
Image 11: Quote from YouTube
Image 11: Quote from YouTube

Jump to position 1 Jump to position 2 Jump to position 3 Jump to position 4 Jump to position 5 Jump to position 6

For everyone: More natural and intuitive interactions

In Gemini Live and Search Live, the 3.1 Flash Live model delivers more helpful and natural responses, whether you’re asking quick daily questions or engaging in more complex conversations.

With the 3.1 Flash Live model under the hood, Gemini Live delivers faster responses compared to the previous model and it can follow the thread of your conversation for twice as long, keeping your train of thought intact during longer brainstorms.

Sorry, your browser doesn't support embedded videos, but don't worry, you can download it and watch it with your favorite video player!

3.1 Flash Live makes Gemini Live faster and more helpful

3.1 Flash Live is also inherently multilingual, which enables this week’s global expansion of Search Live. With this launch, people in more than 200 countries and territories can now have real-time, multimodal conversations with Search in their preferred language.

Sorry, your browser doesn't support embedded videos, but don't worry, you can download it and watch it with your favorite video player!

Get real-time troubleshooting help using 3.1 Flash Live in Search Live

Image 12: Search Live YouTube video
Image 12: Search Live YouTube video

00:00

Search Live is enabling real-time multimodal conversations in more languages

Try Gemini 3.1 Flash Live

All audio generated by 3.1 Flash Live is watermarked with SynthID. This imperceptible watermark is interwoven directly into the audio output, allowing the reliable detection of AI-generated content to help prevent misinformation. For more information on our approach to safety and responsibility, see the model card.

Experience the naturalness and reliability of 3.1 Flash Live, starting today. We look forward to seeing how you interact and build with it.

Image 13
Image 13
Image 14
Image 14
Image 15
Image 15
Image 16
Image 16

Get more stories from Google in your inbox.Get more stories from Google in your inbox.

Email address

Your information will be used in accordance with Google's privacy policy.

Subscribe

Done. Just one step more.

Check your inbox to confirm your subscription.

You are already subscribed to our newsletter.

You can also subscribe with a different email address .

POSTED IN:

Related stories

![Image 17 Gemini models #### Gemini 3.1 Flash TTS: the next generation of expressive AI speech By Vilobh Meshram & Max Gubin Apr 15, 2026](https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-1-flash-tts/)

![Image 18 Chrome #### Turn your best AI prompts into one-click tools in Chrome By Hafsah Ismail Apr 14, 2026](https://blog.google/products-and-platforms/products/chrome/skills-in-chrome/)

![Image 19 Creating opportunity #### Bringing people together at AI for the Economy Forum By James Manyika Apr 14, 2026](https://blog.google/company-news/outreach-and-initiatives/creating-opportunity/ai-economy-forum/)

![Image 20 Google Workspace #### Create, edit and share videos at no cost in Google Vids By David Nachum Apr 02, 2026](https://blog.google/products-and-platforms/products/workspace/google-vids-updates-lyria-veo/)

![Image 21 Developer tools #### Gemma 4: Byte for byte, the most capable open models By Clement Farabet & Olivier Lacombe Apr 02, 2026](https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/)

![Image 22 Developer tools #### New ways to balance cost and reliability in the Gemini API By Lucia Loher & Hussein Hassan Harrirou Apr 02, 2026](https://blog.google/innovation-and-ai/technology/developers-tools/introducing-flex-and-priority-inference/)

.

Jump to position 1 Jump to position 2 Jump to position 3 Jump to position 4 Jump to position 5 Jump to position 6

Image 23
Image 23

Let’s stay in touch. Get the latest news from Google in your inbox.

SubscribeNo thanks

Survey

Help us improve The Keyword with a one-question survey

Yes No

This survey is anonymous. All responses will be aggregated and used only for analysis to improve our services.

Did this article provide the level of detail you were looking for?

Yes, I got what I needed No, I wanted more technical depth No, I wanted a simpler overview I was looking for something else entirely

✅ Thank you!

Follow Us

  • [](https://www.instagram.com/google/)
  • [](https://twitter.com/google)
  • [](https://www.youtube.com/google)
  • [](https://www.facebook.com/Google)
  • [](https://www.linkedin.com/company/google)

[](https://www.google.com/)

*