Google AI developer relations lead Logan Kilpatrick announced on April 15 the release of Gemini 3.1 Flash TTS—the latest text-to-speech model from Google. This model supports 70 languages, fine-grained control at the level of scene direction and speakers, and audio tags. It is now available for use in the audio playground in Google AI Studio and in the Gemini API.
Four core features
Gemini 3.1 Flash TTS comes with four notable upgrades compared with its predecessor:
Scene Direction — You can set a context for the voice, such as “speaking softly in a noisy café” or “excitedly announcing good news,” and the model will adjust tone, speaking pace, and emotion based on the scene
Speaker-Level Specificity — In multi-role conversations, you can set different voice characteristics for each character
Audio Tags — Supports inserting sound-effect instructions into text to control details like pauses and tone changes
Support for 70 languages — Significantly expands multilingual coverage, including Chinese
More natural, more expressive voices
Google emphasized improvements in voice naturalness with this model. Traditional TTS models are often criticized for output that “sounds like AI.” Gemini 3.1 Flash TTS aims to narrow the gap with human speech through richer prosody variations and emotional expression. Kilpatrick noted that progress from Gemini 2.5 to 3.1 is “very significant.”
How developers can use it
Developers can use it in two ways:
Google AI Studio Audio Playground — Test and preview voice effects directly in the web interface
Gemini API — Integrate into applications for scenarios such as voice assistants, audiobooks, automatic Podcast generation, and multilingual customer service
Gemini product line keeps expanding
Flash TTS is part of the recent flurry of releases in the Gemini 3.1 series. Previously, Google rolled out Gemini Robotics ER 1.6 (robot vision reasoning), Tab Tab Tab (Vibe Coding prompt completion), and design preview features. Google is expanding Gemini from a “chat model” into a full-modal AI platform spanning text, speech, vision, and robotics.
This article Google releases Gemini 3.1 Flash TTS: Supports 70 languages and scene direction, for more natural AI voices first appeared on Liannews ABMedia.
Disclaimer: The information on this page may come from third parties and does not represent the views or opinions of Gate. The content displayed on this page is for reference only and does not constitute any financial, investment, or legal advice. Gate does not guarantee the accuracy or completeness of the information and shall not be liable for any losses arising from the use of this information. Virtual asset investments carry high risks and are subject to significant price volatility. You may lose all of your invested principal. Please fully understand the relevant risks and make prudent decisions based on your own financial situation and risk tolerance. For details, please refer to
Disclaimer.
Related Articles
Honor's Lightning Robot Wins Beijing 2026 Humanoid Robot Half Marathon with 50:26 Finish
Honor's "Lightning" humanoid robot set a new record at the 2026 Beijing Yizhuang Humanoid Robot Half Marathon, completing the race in 50 minutes and 26 seconds, exceeding the human world record.
GateNews1h ago
Meta Stock Rises 1.73% as Company Plans 8,000-Job Layoff Starting May 20
Meta Platforms plans to cut about 8,000 jobs, or 10% of its workforce, starting May 20, despite rising stock prices. The company, with over $200 billion in revenue, is focusing on AI investments amid significant restructuring, aligning with industry trends of layoffs.
GateNews9h ago
Google’s annual report says Gemini achieves millisecond interception, blocking 99% of scam ads
The article discusses how Google strengthens ad safety through its generative AI system, Gemini. The report shows that the speed at which it blocks noncompliant ads has been reduced to milliseconds, with a blocking rate of 99%. Last year, Google removed 8.3 billion ad listings and suspended 24.9 million accounts, indicating a significant rise in the number of scam ads. Experts point out that this is a contest between AI and AI, and that in the future there will still be challenges in dealing with both legal and illegal activities brought about by AI.
ChainNewsAbmedia11h ago
Ethereum Co-founder Lubin: AI Will Be Critical Turning Point for Crypto, But Tech Giant Monopoly Poses Systemic Risk
Ethereum co-founder Joseph Lubin emphasized the transformative potential of AI for the cryptocurrency sector while cautioning against the risks of centralization among tech giants. He envisions AI-driven autonomous transactions on blockchain and highlights the convergence of traditional finance with DeFi.
GateNews13h ago
Elon Musk Pushes 'Universal High Income' Checks as Ultimate Solution for AI Unemployment
Elon Musk advocates for a Universal High Income to combat AI-induced unemployment, envisioning a future with ample goods and zero inflation. In contrast, experts like Sam Altman raise concerns about job loss and propose protective measures for workers.
Coinpedia13h ago
DeepSeek Reportedly Launches First External Fundraising Round, Targets $10B+ Valuation and $300M+
DeepSeek, a Chinese AI startup, is negotiating its first external funding round, aiming for at least $300 million at a $10 billion valuation. Despite previous rejections of investment offers, its fundraising discussions are now reportedly underway.
GateNews14h ago
