Gemini 3.8 Flash TTS
Shape every line of speech with Gemini 3.8 Flash TTS — craft custom voices, stage two-person scenes and publish audio in 130 languages, right in your browser.
Support
Pro AI Tools
Explore elite tools including the Suno AI Music Generator.
MiniMax H3
MiniMax H3 AI Video Generator
Seedance 2.5
The Future of AI Video Is Here.

Seedance 2.0
The Future of AI Video Is Here.
FLUX 3 Video Generator

Kling 3.0
Next-Gen AI Video Generator
Grok Video Generator
Create Videos from Text or Images with AI
MiniMax H3 video generator
MiniMax H3 AI Video Generator

Seedance 2.0
The Future of AI Video Is Here.

What Makes Gemini 3.8 Flash TTS a Different Kind of Voice Tool
Launched on September 23, 2026, Google's Gemini TTS family now pairs an expressive creative model with a budget-friendly engine tuned for bulk audio.
- Two Models, Two Very Different JobsGemini 3.8 Flash TTS handles nuanced, character-rich performance, while Flash-Lite TTS keeps costs low when you need thousands of clips fast.
- Style Instructions Instead of Fixed PresetsPer-turn guidance, structured speech metadata and inline vocal tags let you set mood, tempo, emphasis and accent without re-recording anything.
- Build a Voice or Clone One With ConsentType a plain-language description to invent a voice, or mirror a real speaker using a reference clip paired with a matching consent file.
How to Get Clean Results From Gemini 3.8 Flash TTS
Four practical habits that keep your transcript readable and let performance metadata do the heavy lifting.
Inside the Gemini 3.8 Flash TTS Toolkit
Here is a closer look at what this creative-tier Gemini model delivers, from hands-on performance control to wide language coverage.
Line-Level Performance Control
Assign a style to each turn and drop in laughs, sighs, coughs, breaths or pauses — it feels like coaching an actor, not selecting a stock voice.
Voices Designed With Plain Language
Describe age, personality, accent, vocal texture and role in a prompt, with 2,000+ ready-made voices also available through the Voices endpoint.
Voice Cloning Behind a Consent Gate
You need a clean reference sample plus a consent recording from the same adult speaker, and output carries SynthID watermarks and C2PA credentials.
Built-In Two-Speaker Scenes
Podcasts, teaching dialogues, product walkthroughs and game scenes can be scripted as a conversation instead of stitched line by line.
Consistent Voice Across Long Recordings
Google reports steady identity, timbre, loudness and room tone through multi-minute narration and extended back-and-forth dialogue.
130 Languages With Regional Nuance
Flash TTS supports 130 languages versus 101 for Flash-Lite, including regional accents, minority dialects and IPA pronunciation overrides.
Gemini 3.8 Flash TTS — Your Questions Answered
Quick answers about Gemini 3.8 Flash TTS pricing, which model tier to pick, benchmark standings and built-in safeguards.
What does Gemini 3.8 Flash TTS cost per audio minute?
Roughly 1.35 cents per minute, based on launch pricing of $0.50 per million input tokens and $9 per million output tokens.
When should I choose Flash TTS over Flash-Lite TTS?
Pick Flash TTS when acting detail and long-form narration matter; choose Flash-Lite TTS for high-volume jobs and fast turnaround.
How does it score against rival voice models?
Google cites 71.4 on Hume's Voice Design Benchmark, and Voice Arena ranks it second with an Elo of 1,260.
What changed from Gemini 3.1 Flash TTS Preview?
Flash-Lite TTS takes over from the 3.1 preview and drops audio output pricing from $20 to $6 per million tokens.
Can I clone a voice, and what protections exist?
Cloning requires both a reference clip and a consent recording from the same adult speaker before any audio is produced.
Why is the model reading my stage directions aloud?
Everything in the input box is treated as spoken text, so move long-lasting directions into the speech metadata instead.
Bring Gemini 3.8 Flash TTS to Your Next Voice Project
Try both tiers in the Gemini API or Google AI Studio — moving between Gemini 3.8 Flash TTS and Flash-Lite TTS only means changing the model identifier. Test batch versus priority inference first, then lock in your production budget.
