Visibl

What it does

Visibl is an iOS app that transforms audiobooks into immersive videobooks, turning your listening experience into a visual journey. As you listen, Visibl dynamically generates images in real-time, allowing users to see a unique visual interpretation of the audiobook they are enjoying. The app doesn't just create static images; it enables users to influence and guide the visuals, making each journey through a book highly personalized.

Visibl leverages the power of the Gemini API in three key ways:

- Audio Transcription: The app transcribes the audiobook audio into text, ensuring accurate representation of the content.

- Named Entity Recognition (NER): This allows the app to identify and focus on key characters, places, and objects within the text, which are crucial for generating contextually relevant visuals (Gemini 1.5 Pro)

- Image Prompt Generation: Using the insights from transcription and NER, the app generates detailed and personalized image prompts, which are then used by diffusion models to create the visuals in real-time. (Gemini 1.5 Pro)

This combination of Gemini API features ensures that Visibl not only provides a novel way to experience audiobooks but also tailors the experience uniquely to each user.