Abstract

This document describes techniques that enable a computing device (e.g., a smartphone, smart watch, wearable device, augmented reality glasses, vehicle infotainment system, etc.) to generate an adaptive audio digest. The system may use one or more artificial intelligence (AI) models, (e.g., a large language model (LLM), predictive LLM, generative model, machine learning model, etc.) to dynamically generate and personalize the adaptive audio digest (e.g., a personalized audio broadcast). With explicit user consent, AI models may pull content from various data sources and/or applications (e.g., calendar items, news preferences, music/podcast subscriptions, reminders, etc.). The AI may intelligently arrange this content, perhaps starting with a summary of the user’s first meeting, followed by a synopsis of top news articles, and then a few of the user’s favorite songs. If the predicted length of the audio digest changes (e.g., due to changes in traffic conditions, various delays, etc.), the AI models automatically adjust the adaptive audio digest by adding, removing, or shortening content segments based on the changes in the predicted length. In this way, the technology may provide a hands-free and highly personalized listening experience that makes travel time more productive and enjoyable, concluding as the user arrives at their destination.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.

Download

Share

COinS