Gemini 3.8 Flash TTS

Audio modelCard summarized

Flagship creative text-to-speech model

● Re-checked 12 Oct 2026 · 49 facts · 2 corrections · 1 unconfirmedSource: Gemini 3.8 Audio - Model Card

Use it via API

$0.00022 / sec

Code example Compare

Where to call it

Gemini APIgemini-3.8-flash-tts
Google Vertex AIgemini-3.8-flash-tts

Supports

Streaming
Batch API

Full feature support

Good for

Studio-grade creative work and long-form narration
Complex multi-speaker dialogue
Voice design from prompts and voice replication

Overview

Google · 01Flash

3.8 Flash TTS

Flagship creative text-to-speech model

Typical price

$0.00022 / sec

Audio output, standard

Max length: —
Languages: 130+ languages (auto-detected)
Modes: text-to-speech, multi-speaker (2 voices)
Released: 23 Sep 2026

One card covers Gemini 3.8 Live, Live Extended Thinking, Flash TTS and Flash-Lite TTS ("Gemini 3.8 Audio"), an addition to the Gemini 3 series based on Gemini 3 Pro. Architecture, training data and hardware are those of Gemini 3 Pro; the card reports no benchmark scores. Model information

gemini-3.8-flash-tts

Typical price: $0.00022 / sec
Languages: 130+ languages (auto-detected)
Weights: Closed
Released: 23 Sep 2026

text-to-speech
multi-speaker (2 voices)
streaming
voice design
voice replication
inline style tags