Security and responsibility
We have proactively assessed potential risks at each stage of the development process for these native audio features, using what we have learned to enlighten our mitigation strategies. We validate these measures through rigorous assessments of internal and external security, including red team for responsible deployment. In addition, all the audio outputs of our models are integrated SyntheticOur watermark technology, to ensure transparency by making its audio generated by the identifiable AI.
Native audio capacities for developers
We provide native audio outputs to Gemini 2.5 models, giving developers new capacity to create richer and more interactive applications via the Gemini API Google Ai Studio Or Vertex ai.
To start exploring, developers can try the native audio dialog box with Gemini 2.5 Flash seen in Google AI Studio flow tongue. The generation of controllable speeches (TTS) is available in preview for Gemini 2.5 Pro and Flash by selecting the generation of speeches in the generate media Tab in Google AI Studio.