Model / API profile · Checked September 27, 2026

Gemini 3.8 Flash TTS and Flash-Lite TTS

Gemini 3.8 Flash TTS and Flash-Lite TTS turn text into audio through Google AI Studio and the Gemini API. This is a TTS model/API profile, separate from the general Gemini assistant.

Release: 2026-09-23. Source-verified; not independently benchmarked.

Video Audio · Official starting point

Choose Flash or Flash-Lite

Flash targets expressive voice and character direction; Flash-Lite targets higher-volume speech workloads. Both support two-speaker scripts and controls for delivery, pacing and dialect. Google describes voice design and consent-based replication; remixing is described as coming soon. The model pages list 130 languages for Flash and 101 for Flash-Lite. Availability of individual voices and replication varies by region.

Who it is for

Developers building narration, dialogue or voice-content workflows who can prepare scripts, configure speakers and review the resulting audio. Use a packaged voice editor instead if you need timeline editing, asset management and a finished production interface.

API access and pricing

Use model IDs gemini-3.8-flash-tts or gemini-3.8-flash-lite-tts. A limited free tier is listed. Standard paid rates per million tokens through December 31, 2026 are $0.50 text input and $9 audio output for Flash, or $0.50 and $6 for Flash-Lite. Listed January 1, 2027 rates rise to $1/$18 and $1/$12 respectively. These are token prices, not monthly subscriptions or per-minute guarantees. Check quota and billing eligibility before integrating.

Capability boundaries and first check

The models accept text and return audio; they are not ASR or the Live API. Model documentation lists 8,192 input and 16,384 output tokens. Put delivery instructions in the documented speech metadata rather than assuming inline directions will stay unspoken. Start with a short two-speaker script, then listen for speaker consistency, pronunciation and unwanted spoken instructions. Replication requires the appropriate consent and is unavailable in some regions; confirm current restrictions in Google’s announcement.

Official sources

Checked September 27, 2026. Access and prices can change; follow the linked provider documentation before use.

Related workflows

General Gemini profile · Compare voice generators