text-to-speech gemini-3.8 ai-gateway voice-generation vercel

Gemini 3.8 text-to-speech models launch on AI Gateway

Two new Google TTS models (Flash-Lite for high-volume, Flash for character control) available via unified API with per-request cost tracking.

Summary

Consolidates speech generation into your existing AI Gateway setup, eliminating separate provider integrations. Flash variant enables natural-language voice design—accents, acting cues, conversational reactions—without manual audio engineering.

Why it matters

Implementation verdict

Replaces direct Google TTS calls or separate Elevenlabs integrations. Requires: API key, Node.js 18+, one function call with model ID. Ready now—playground and quickstart available. Worth trying if you're already on AI Gateway; marginal friction if not.

Sources

1.Both models take text and generate speech in more than 100 languages
2.google/gemini-3.8-flash-lite-tts is suited to high-volume speech generation, with controls for tone, pacing, and line-by-line delivery
3.google/gemini-3.8-flash-tts adds voice and character design through natural-language prompts, including acting cues, accents, and conversational reactions
4.AI Gateway provides one API for speech generation alongside your other models, with usage and cost tracking for each request

Dev Signal

Get briefs like this in your inbox — free, every weekday.

100+ sources compressed into one 4-minute read. Ranked, cited, implementation-ready.

Read the full issue →All briefs