Safety and responsibility
We actively assessed potential risks at every stage of the development process for these native audio features, using what we learned to develop our mitigation strategies. We validate these measures through rigorous internal and external security assessments, including comprehensive ones red team for responsible implementation. Additionally, all audio outputs of our models are equipped with SynthID, our watermarking technology, to ensure transparency by enabling the identification of AI-generated audio.
Native audio capabilities for developers
We're introducing native audio output to Gemini 2.5 models, giving developers new opportunities to create richer, more interactive applications via the Gemini API in Google Artificial Intelligence Studio Or Apex AI.
To start exploring, developers can try native audio dialogs with Gemini 2.5 Flash preview in Google AI Studio stream strap. Controlled speech generation (TTS) is available in preview for both Gemini 2.5 Pro and Flash after selecting speech generation in the menu generate media in Google AI Studio.
