← All models

Gemini 3.8 Flash TTS

Published Publisher Googlegemini-3.8-flash-tts

Overview

Gemini 3.8 Flash TTS is Google's flagship creative text-to-speech model. It converts text into expressive audio, with an emphasis on voice fidelity, regional accents and consistent long-form, multi-turn speech. It supports single- and multi-speaker generation, delivery controls, custom voice design and consent-verified voice replication. Google lists 130 supported languages for the Gemini API. (aistudio.google.com)

Uses and limitations

Suitable uses include audiobooks, podcasts, studio narration, character voices and multi-speaker dialogue. The Gemini API model accepts text and outputs audio, with limits of 8,192 input tokens and 16,384 output tokens. Context caching, function calling and the Live API are not supported. Text intended as an instruction may be read aloud; Google recommends putting delivery and speaker directions in structured metadata. Voice replication requires consent verification. (aistudio.google.com)

Release

The Gemini API release notes list general availability on September 22, 2026. Google's product announcement is dated September 23, 2026. (aistudio.google.com)

Gemini API pricing

Official paid-tier rates are in USD per 1 million tokens, with text input and audio output billed separately. Rates below were checked on November 16, 2026; they apply to the Gemini API, not necessarily other Google platforms. (aistudio.google.com)

Service tier Through December 31, 2026: input / output From January 1, 2027: input / output
Standard $0.50 / $9.00 $1.00 / $18.00
Batch $0.25 / $4.50 $0.50 / $9.00
Flex $0.25 / $4.50 $0.50 / $9.00
Priority $0.90 / $16.20 $1.80 / $32.40

The pricing page lists free input and output for the Standard and Priority free tiers; free-tier Batch and Flex are unavailable. Google states that audio output corresponds to 25 tokens per second. (aistudio.google.com)

Official sources