Google's new Flash TTS models let you design AI voices from scratch using text descriptions
Google just shipped text-to-speech that builds voices from descriptions—no cloning needed. Over 100 languages, stage directions, two-voice dialogue from one script.
Why it matters
Google's new Flash TTS models expand the accessibility and creative control of AI voice generation for developers and creators, moving beyond simple speech synthesis to scriptable, multi-voice dialogue production at scale.
The key facts
10 to knowGemini 3.8 Flash TTS and Flash-Lite TTS announced
Supports 100+ languages
Create new voices from text descriptions (generative voice design)
Voice cloning from 30-second audio sample
Stage directions per line support
Two-voice dialogue generation from single script
Two new models: Gemini 3.8 Flash TTS and Flash-Lite TTS
Support for 100+ languages
Voice design from text descriptions (e.g., 'warm, authoritative female voice')
Stage directions for individual lines
Go to the source
The Decoderthe-decoder.com
Publisher excerpt: Google is introducing two new text-to-speech models, Gemini 3.8 Flash TTS and Flash-Lite TTS, which support more than 100 languages. Flash TTS can create new voices from text descriptions, and both models let users add stage directions to individual lines and generate two-voice dialogue from a…