Gemini 3.8 Flash TTS
Shape every line of your script with Gemini 3.8 Flash TTS — direct emotion, stage two speakers and render 130 languages from a single prompt.
Support
Pro AI Tools
Explore elite tools including the Suno AI Music Generator.
Nano Banana2
Best Image Generator
Kling3
Next-Gen AI Video Generator
Veo3
Create Stunning Videos with Veo3.1
Seedance2.0
The Future of AI Video Is Here.
Kling Motion Control
Turn reference images into amazing motion videos in minutes
Suno AI Music Generator
Create Professional Music with AI
Nano Banana
Advanced AI Image Generator
AI Image Editor
AI Photo Editor Online
Gemini 3.8 Flash TTS: A Creative Engine for Spoken Audio
The September 2026 rollout divides Google's TTS line into a nuance-first creator model and a lean, high-volume engine for mass audio.
- A Single Launch, Two Very Different JobsGemini 3.8 Flash TTS handles nuance-driven work, while its lighter sibling keeps bulk narration fast and inexpensive.
- You Direct the Delivery Instead of Choosing PresetsPer-turn style notes, structured speech metadata and inline vocal cues control tone, tempo, feeling and accent.
- Invent a Voice or Replicate One With ConsentSketch a persona in plain words, or mirror a real person using a reference sample paired with a consent recording.
How to Prompt Gemini 3.8 Flash TTS for Clean Takes
Four habits that keep your script readable and let the direction metadata do the acting.
What Gemini 3.8 Flash TTS Can Actually Do
Direction control, voice design and global language reach — a closer look at the model's real strengths.
Line-Level Performance Control
Style notes per turn, plus embedded laughs, sighs, breaths and beats, feel more like coaching a performer than picking a stock voice.
Design Voices With Plain Words
Describe age, temperament, accent, timbre and role in a sentence, drawing on 2,000+ ready-made voices from the Voices endpoint.
Cloning Locked Behind Consent
You need a clear sample and a consent clip from the same adult voice, all wrapped in SynthID watermarking and C2PA provenance.
Built-In Two-Person Scenes
Podcasts, lesson dialogues, walkthroughs and game banter come out as one continuous take — no stitching lines together by hand.
Consistent Voice Across Long Reads
Google reports steady identity, timbre, loudness and room tone through minutes of narration and drawn-out conversations.
Global Reach With Regional Flavor
The expressive tier handles 130 languages versus 101 on the lighter model, including regional accents, minority dialects and IPA overrides.
Gemini 3.8 Flash TTS: Your Questions Answered
Quick answers about cost, choosing between the tiers, benchmark scores and the safety rules around voice cloning.
What does Gemini 3.8 Flash TTS charge per minute of audio?
Roughly 1.35 cents for every audio minute, based on launch pricing of $0.50 per million input tokens and $9 per million output tokens.
Which tier fits my project — Flash TTS or Flash-Lite TTS?
Choose the expressive model when acting nuance and long narration matter; pick the Lite version when you need volume and minimal delay.
How does it score next to rival voice models?
Google cites a 71.4 score on Hume's Voice Design Benchmark, while Voice Arena ranks it second with 1,260 Elo.
What changed from the Gemini 3.1 Flash TTS Preview?
The Lite tier supersedes that preview and drops audio output pricing from $20 down to $6 per million tokens.
Can I clone a voice, and what safeguards apply?
Yes, but only with a reference sample and a consent recording supplied by the same adult speaker.
Why is the model speaking my stage directions aloud?
Everything you submit is treated as the script itself, so shift any lasting direction into the metadata fields.
Point Gemini 3.8 Flash TTS at Your Next Script
Try both tiers inside the Gemini API or Google AI Studio — moving between Gemini 3.8 Flash TTS and its Lite sibling means swapping one model identifier. Weigh batch against priority inference before locking in a production budget.