Gemini 3.8 text-to-speech models now available on AI Gateway
Google's Gemini 3.8 TTS models hit AI Gateway—100+ languages, character design, dual-speaker dialogue. Practitioners now route speech generation through one unified API.
Why it matters
Google expands its text-to-speech capability tier and makes it accessible via a unified API gateway, giving practitioners easier integration and cost tracking for multilingual voice generation at scale.
The key facts
11 to knowTwo models: Gemini 3.8 Flash-Lite TTS (high-volume, tone/pacing controls) and Gemini 3.8 Flash TTS (character design, accents, acting cues via natural language)
Supports 100+ languages with long-form narration and two-speaker dialogue
Available on AI Gateway with unified API, usage tracking, cost per-request visibility, and routing rules
Flash-Lite TTS optimized for high-volume generation; Flash TTS optimized for character and voice control
Includes playgrounds and quickstart guides for immediate testing
Published Sept 23, 2026
Gemini 3.8 Flash-Lite TTS and Gemini 3.8 Flash TTS both available
Support for 100+ languages
Features: long-form narration, tone/pacing control, two-speaker dialogue, character design via natural-language prompts
Available via Vercel AI Gateway with unified API, usage tracking, and routing rules
Flash-Lite TTS optimized for high-volume generation; Flash TTS supports voice/character design
The story so far
Earlier coverage of this storyline
- Google’s new speech model Gemini 3.8 Live supports real-time reasoningSiliconAngle
- This story
Go to the source
Vercel Blogvercel.com
Publisher excerpt: and from Google are now available on .Gemini 3.8 Flash-Lite TTSGemini 3.8 Flash TTSAI Gateway Both models take text and generate speech in more than 100 languages. They support long-form narration, control over delivery, and two-speaker dialogue. Generate and listen to speech in the or the . For…