TECHNOLOGY
Google Unveils Gemini 3.8 Flash TTS Models for Custom AI Voice Generation
Google has introduced Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS, which it describes as its most expressive text-to-speech models yet, rolling out across its AI Studio, API and enterprise tools.
Image: IntraGoals Media · Uploaded by IntraGoals — usage rights confirmed
Google has announced two new text-to-speech models, Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS, which the company describes as its most expressive audio generation models to date. The announcement, first surfaced on Hacker News and attributed to a director of research science speaking on behalf of the Gemini Audio Team, positions the models as a significant step forward in Google's push into synthetic voice technology.
According to Google, the new models allow users to generate custom character voices and direct dialogue for specific scenes, with control over qualities such as accent and emotional tone. The company says the tools are aimed at creators working on audiobooks, video games and podcasts who want voices that sound convincingly human.
The models are being made available across several Google products, including Google AI Studio, the Gemini API, Gemini Enterprise, Gemini Notebook and Google Vids. That distribution suggests Google intends the technology to serve both individual developers experimenting through its API and larger enterprise customers building voice features into their own products.
Google also says it has built in safety features intended to guard against misuse of the voice-generation tools, though the company has not yet detailed exactly what those safeguards entail. As synthetic voice technology becomes more realistic, questions about consent, impersonation and misinformation have followed closely behind, and safeguards of this kind are likely to draw scrutiny once the models are more widely tested.
Details beyond Google's own description remain limited. The original source material offers only a brief overview of the models' capabilities, and specifics such as pricing, availability timelines and technical benchmarks have not been independently confirmed. Readers should treat the “3.8” naming and full feature set as preliminary until Google publishes fuller documentation or the claims are verified against the company's official channels.