Gemini 3.8 Flash TTS goes live on Runware with 130+ languages and two-speaker dialogue

aziz4ai · x · 2026-11-21

Google DeepMind's Gemini 3.8 Flash TTS is now available via Runware's API (MCP supported). Key upgrades: delivery directions for tone/pace/emotion plus inline <laugh>, <sigh> and <short pause> tags; two-speaker dialogue in one request with 30 prebuilt voices; language auto-detection expanded from 70+ to 130+; up to 4,000 characters per request. Pricing: $1.00/1M input tokens and $18.00/1M output tokens. Gemini 3.1 Flash TTS deactivates Dec 18 — switch your model ID to google:[email protected].

Original post →

More from Multimodal

Multimodal channel →