We live in the era of ubiquitous sensing and computing. More and more data is being collected and processed from devices, sensors and systems. This opens up opportunities to create insights from these data that could help in gaining better understanding of the source (e.g. human who produces the data). This can be useful in a wide range of domains,
and especially in the area of personal health. Such insights could help to improve people lifestyle, change their behavior and address important societal challenges such as obesity or cardio-vascular diseases. In this experiment, we propose to apply the deep learning-based, low-energy Insight Generator to a fitness app. Consumers automatically collect data about their activity and nutrition intake using a smart watch and a mobile
phone with the existing 5M ICT FitSprite app. We add the Insight Generator to the app to identify actionable insights, verbalize them in a readable text format and present them to the consumer, his coach or doctor.