Google Releases Gemini 3.8 Flash TTS and Flash-Lite TTS With Prompt-Based Voice Design
Google's new Flash TTS tops voice-design benchmarks with prompt-based synthesis across 100+ languages — reshaping how agents and dubbing services generate natural speech.
Why it matters
Google ships a measurable step forward in text-to-speech capability (benchmark leader) with a cost tier (Flash-Lite) for production voice agents. Practitioners deploying voice-heavy agentic systems now have a new tier to evaluate; enthusiasts see the frontier labs competing on multimodal capability depth.
The key facts
5 to knowGemini 3.8 Flash TTS and Flash-Lite TTS released via Gemini API and AI Studio
Flash TTS ranks #1 on Hume AI's Voice Design Benchmark (score: 71.4)
Supports 100+ languages with prompt-based voice design
Flash-Lite TTS optimized for high-volume dubbing and voice agents at lower cost
Available now through public APIs
The story so far
Earlier coverage of this storyline
- Gemini 3.8 TTS PlaygroundSimon Willison
- Google's new Flash TTS models let you design AI voices from scratch using text descriptionsThe Decoder
- This story
Go to the source
MarkTechPostmarktechpost.com
Publisher excerpt: Google has released Gemini 3.8 Flash TTS and Flash-Lite TTS, 2 new text-to-speech models available now through the Gemini API and Google AI Studio. Flash TTS designs new voices from natural language prompts across 100+ languages. It ranks #1 on Hume AI's Voice Design Benchmark with a score of 71.4.…