Security and responsibility
We proactively assessed the potential risk at every stage of the development process for these native sound functions, using what we have learned to inform about our relief strategies. We check these funds through strict internal and external safety assessments, including comprehensive ones Red teams For responsible implementation. In addition, all audio outputs from our models are set in Synthid, our water marking technology to ensure transparency by making sound generated by AI.
Native audio possibilities for programmers
We introduce native audio outputs to the Gemini 2.5 models, giving programmers new opportunities to build richer, more interactive applications through API Gemini Google to learn Or Vertex AI.
To start exploration, programmers can try the native audio dialog with Flash Gemini 2.5 preview in Google AI Studio stream strap. Controlled speech generation (TTS) is available in view for both Gemini 2.5 Pro and Flash, choosing speech generation in media Google AI Studio tab.
