Gemini 3.8 Flash TTS
Type a script and Gemini 3.8 Flash TTS performs it — craft a voice from a description, run two-role dialogue, and publish in 130 languages.
Support
Pro AI Tools
Explore elite tools including the Suno AI Music Generator.
Lyrics to Song
Turn your lyrics into professional songs with AI
Suno 5.5
Create Professional Music with AI
AI Music Generator
Create music from videos with AI
AI Song Generator
Create songs from videos with AI
Suno AI Music Generator
Create Professional Music with AI
Seedance2.0
The Future of AI Video Is Here.
Happy Horse 1.0
Veo3.1
Create Stunning Videos with Veo3.1
Gemini 3.8 Flash TTS in Context: Two Tiers, Two Workloads
Launched on 23 September 2026, this release divides Google's speech family into an expressive flagship and a fast, budget-friendly engine for bulk rendering.
- Two Tiers From a Single LaunchThe flagship handles nuanced, performance-led audio, while Flash-Lite TTS keeps spend down when volume matters more than nuance.
- Style Controls That Go Beyond PresetsPer-turn instructions, structured speech metadata and inline vocal cues let you shape tone, timing, emotion and accent.
- Describe a Voice, or Clone One With ConsentBuild a persona from a written description, or recreate a real speaker using a reference clip plus a matching consent recording.
How to Write Prompts for Gemini 3.8 Flash TTS
Four habits that keep your script clean while the performance metadata does the heavy lifting.
What Gemini 3.8 Flash TTS Can Actually Do
From acting cues to multilingual coverage, these are the capabilities that separate the flagship tier from the budget one.
Performance Cues at the Turn Level
Set a style for each turn and drop in laughs, sighs, coughs, breaths or pauses — it feels like coaching an actor rather than choosing a stock voice.
Build a Voice From a Written Description
Prompts can specify age range, personality, accent, vocal texture and role, with more than 2,000 production voices available through the Voices endpoint.
Cloning That Requires Documented Consent
You need a clean reference recording plus a consent recording from the same adult speaker, and every output carries SynthID watermarking and C2PA credentials.
Ready-Made Two-Role Conversations
Write the exchange once for podcasts, lessons, product walkthroughs or game scenes — no stitching separate takes together afterwards.
Steady Voice Across Long Recordings
Google documents consistent identity, timbre, loudness and room tone across multi-minute narration and extended back-and-forth dialogue.
Wide Language and Accent Coverage
The flagship tier handles 130 languages against 101 on Flash-Lite, including regional accents, minority dialects and IPA pronunciation overrides.
Gemini 3.8 Flash TTS: Your Questions Answered
Quick answers on cost per minute, picking a tier, benchmark scores and the safeguards around voice cloning.
What does it 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.
Which tier should I pick — Flash or Flash-Lite?
Choose the flagship for acting nuance and long-form work; pick Flash-Lite when you need bulk volume and low latency.
How does it score against rival voice models?
Google cites 71.4 on Hume's Voice Design Benchmark, and Voice Arena places it in second position with 1,260 Elo.
What changed from the 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 checks are in place?
Cloning needs a reference clip and a consent recording from the same adult speaker before any audio can be produced.
Why is the model reading my stage directions aloud?
Input is treated as a verbatim transcript, so move any long-running directions into the speech metadata fields.
Hear Your Own Script in Gemini 3.8 Flash TTS
Try both tiers in the Gemini API or Google AI Studio — moving between Gemini 3.8 Flash TTS and Flash-Lite TTS takes nothing more than swapping a model ID. Compare batch and priority inference before you lock in a production budget.