- Google launched Gemini 3.8 Flash TTS and Flash-Lite TTS for expressive, customizable voice generation
- Gemini 3.8 Flash TTS targets creative uses like gaming, audiobooks, and interactive media
- Users can create custom voices with accents, roles, and detailed performance directions
Google has introduced Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS, two new text-to-speech models designed to bring more expressive, customisable voice generation to creators, developers and businesses.
According to Google, the new models move beyond preset synthetic voices by letting users create bespoke vocal identities, direct dialogue, and control how individual lines are delivered.
The tools are available across Google AI Studio, Gemini API, Gemini Enterprise, Gemini Notebook and Google Vids.
Also Read | Layoffs Are Rising In Tech, But These Skills Are Helping People Unlock Better Pay
The higher-end Gemini 3.8 Flash TTS is designed for creative applications such as gaming, audiobooks, podcasts and interactive media.
Users can generate original voices through natural-language prompts and specify characteristics including accents, roles and other vocal attributes across more than 100 languages and dialects.
Also Read | EU Fines Google 403 Million Euros For Location Data Breach
The model also supports detailed performance direction, including pacing, acting cues, dialect changes and conversational backchanneling. Google said users can access more than 2,000 production-ready voices, while custom voices can be created and saved for repeated use.
For voice replication, the model can recreate a vocal profile from a 30-second audio sample, provided the user has the necessary rights. Google said the process includes consent verification, while generated audio carries SynthID watermarking and C2PA credentials.
The Gemini 3.8 Flash-Lite TTS model is aimed at high-volume applications where cost and scale are more important. Google highlighted uses such as large-scale dubbing, audio production and expressive voice agents.
Both models support long-form audio generation and can stage conversations involving two speakers while maintaining distinct voices and natural turn-taking. Users can also add non-verbal cues such as laughter, sighs and gasps to scripted performances.
Also Read | Googlebook Launched: Price, Features, Specifications, Hardware And Other Details
Google said Gemini 3.8 Flash TTS ranked first on Hume AI's Voice Design Benchmark with a score of 71.4 and also led its accent-modelling evaluation. The two models placed first and second respectively on Hume AI's Overall Quality Index.
Developers can try the models through Google AI Studio, while platforms including Agora, LiveKit, Pipecat and Vercel are integrating the technology through the Gemini API.
Google said the new models are also being adopted by companies such as Figma, HeyGen, Linguana, Wondercraft, 99.co and Ollang for dubbing, localisation and voice-agent applications.
Essential Business Intelligence, Sharp Market Insights, Practical Personal Finance Advice, Daily Fuel, Gold and Silver Prices and Latest Stories — On NDTV Profit.
