Gemini 3.8 Flash TTS and Flash-Lite TTS reach GA

Google has made Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS generally available through the Gemini API. The release also adds the Voices endpoint, /v1beta/voices, for discovering and using prebuilt and custom voices.

Two models for different voice workloads

gemini-3.8-flash-tts is the flagship creative model. Google describes it as engineered for studio-grade voice fidelity, nuanced acting, regional dialects, and long-form multi-turn stability.

gemini-3.8-flash-lite-tts is the faster, cost-efficient option. Google positions it as a replacement for gemini-3.1-flash-tts-preview in high-throughput production and real-time voice-agent cascades.

Custom voices and the Voices endpoint

The GA release adds voice design from text prompts, voice replication with consent verification, and access to more than 150 prebuilt and custom voices. The endpoint gives applications a way to query that voice library instead of hard-coding a single voice into every integration.

Voice selection is only one part of a reliable voice agent. Test pronunciation, interruption behavior, latency, consent records, fallback behavior, and how generated audio is disclosed to users before expanding a custom persona to production.

Migration checklist

Teams moving from the preview TTS model should test the new model IDs and confirm account access, quotas, and pricing in the current Gemini API documentation. Compare Flash TTS and Flash-Lite TTS on the same scripts and interaction traces; the most expressive voice is not always the best choice for a high-volume or low-latency workflow.

Sources

• Google AI for Developers — Gemini API release notes (September 22, 2026): https://ai.google.dev/gemini-api/docs/changelog
• Google AI for Developers — Gemini API changelog text: https://ai.google.dev/gemini-api/docs/changelog.md.txt
• Google for Developers on X — official developer updates: https://x.com/googledevs