Google Launches Gemini 3.5 Transcribe and Omni 1.1 Flash for Audio and Video Workflows

Gemini 3.5 Transcribe is a new speech-to-text model that converts raw audio inputs into highly accurate, formatted text while effectively handling background noise and complex industry jargon. [1] Developers can access the model via a Live API for real-time streaming applications, as well as an Interactions API for pre-recorded audio processing tasks. [1]

Google introduced Gemini Omni 1.1 Flash, which brings generative video capabilities to the Gemini API in Google AI Studio. [2] Omni 1.1 Flash allows users to extend videos in 10-second increments up to a cumulative length of 40 seconds. [2] The model also lets users specify starting and ending frames to generate continuous video between two keyframes. [2]

Google is expanding its developer ecosystem with production-ready multimodal tools, offering both sub-second audio transcription and high-resolution video generation. By releasing these capabilities concurrently via the Gemini API, Google enables developers to build complex applications ranging from voice agents to media editing software. This dual release positions Google's AI Studio as a comprehensive hub for both audio and visual generative tasks. What the sources don't address: the pricing structures and specific usage limits for processing large volumes of media through these new APIs.