Gemini 3.8 Flash TTS turns voice generation into your full creative studio. Design original character voices from scratch across 100+ languages, stage two-speaker dialogue, and tailor every performance line by line with natural laughs, sighs, whispers, and more.

Learn more about Gemini 3.8 Flash TTS ↓

@GoogleDeepMind TFW people are looking for Pro and keep getting more Flash variants! 😭

@GoogleDeepMind Any chance of fixing Gemini for Google search?

@GoogleDeepMind the ability to direct every little part of a performance is what makes this so useful!!!

@GoogleDeepMind Two voices and natural sighs: every couple trying to pick a restaurant.

@GoogleDeepMind You can create a game from home with three subscriptions.

@GoogleDeepMind If this thing can actually nail a heated argument between two characters - like, voices overlapping, real tension - I’m in. Most AI voices still sound way too polite. Has anyone tried pushing it with messy dialogue or real emotion yet?

@GoogleDeepMind @GoogleDeepMind @GoogleAI @GeminiApp It is perfect on English and German, but still strange on Cantonese.

@GoogleDeepMind elevenlabs did this 2 years ago bro

@GoogleDeepMind SICK.

@GoogleDeepMind Claude: sonnet 5.5 opus 5.5, openAi: Astra, sol 6, 20 launches tomorrow, and google: TTS

@GoogleDeepMind Gemini ke TTS se ab knight bhi "You shall not pass" bol sakta hai… lekin asli sawaal yeh hai: AI ko character voice design karne do toh human actors ka future kya hoga? Ya phir yeh sirf ek aur tarika hai originality ko kill karne ka?

@GoogleDeepMind Is my Dream to get a member Of Google as a Senior Projektmanager SCM #Geopolitik @sundarpichai

@GoogleDeepMind 逐句调语气挺实用，就看两个人对话时能不能接得自然。

@GoogleDeepMind what if someone paid for your pro plan and never got a pro model for many months now

@GoogleDeepMind A good concept but I don't think I would prefer this for something like Baldur's Gate. For example Astarion's voice.

@GoogleDeepMind Google 推出 Gemini 3.8 Flash TTS，把语音生成做成可设计角色声线、双人对话与逐行情绪控制的创作台，覆盖 100 多种语言。TTS 竞争正在从「像不像真人」挪到「能否按剧本稳定改表演」；可编辑的是笑声、叹息与耳语一类表演单元，而不只是通篇音色。

@GoogleDeepMind Line-by-line control over laughs and sighs is the part that actually matters for audiobooks—wonder how it handles pacing.

@GoogleDeepMind Where the hell is google heading , someone explain me pls ? OpenAI gonna do 20+ launches Claude had lunched opus 5.5 Google is ... ?

@GoogleDeepMind Your whole company is a joke

@GoogleDeepMind 100+ languages with natural laughs and whispers is the game dev unlock. one solo dev can now voice a full rpg party without hiring 6 actors. elevenlabs must be nervous today.

@GoogleDeepMind Granular acoustic nuance across global languages elevates speech synthesis from creative tooling to foundational interface infrastructure.

@GoogleDeepMind We need Gemini Omni 3 and Gemini 4 No more flash model

@GoogleDeepMind 100 languages and a sigh. ours only does hold music.

@GoogleDeepMind Nice but Imagine if Gemini 4 is the model that coded the game/3D scene, character...etc

@GoogleDeepMind for an audiobook maybe, for a game not sure

@GoogleDeepMind Whispers are where demos usually fumble. Show one you can't fake.

@GoogleDeepMind “Natural laughs and sighs” sounds useful right up to the moment one lands in the wrong dramatic beat. Does Gemini expose timing and intensity controls, or is actor direction still guesswork?

@GoogleDeepMind Flash-flash-flash..........

@GoogleDeepMind

@GoogleDeepMind