There are many APIs and products in the industry you can use to generate audio & images, but most of them aren't free. This includes Google AI products like Veo and Imagen... they aren't free either. What if you just want to experiment for free and don't need high quality output? This is the perfect use case for Gemini, a robust and fully-featured LLM (large language model) that has such capabilities. In this session, learn how to use the Gemini API to generate images as well as spoken-voice audio files using the Flash 2.0 Experimental model. Also get a peek at the capabilities of the Flash 2.5 Image (Nano Banana) model!

intended for software developers building genAI apps or guiding the development of vibecoded apps or agents