what is ride-recap fr?

Ride Recap takes hours of raw GoPro video footage from a bike ride, combines it with a Garmin data file that records speed, heart rate, power, and location for every second of the ride, and automatically produces a short 60-second highlight video with that ride data displayed on screen. The author built it because they ride most weekends and end up with hours of footage that nobody, including themselves, wants to sit through, but manually editing a highlight reel every time takes too much effort to bother with. The tool works by scoring every moment of the ride from several angles at once: the Garmin data flags exciting bursts like sprints or climbs, the Strava fitness app can flag well-known popular routes, and an AI vision model watches the actual video frames and rates how visually interesting each one is. An optional tool also lets a person manually mark their favorite moments. All of these signals get combined, favoring moments where multiple sources agree something is worth keeping, and the tool picks a set of clips spread out across the ride rather than clustering them all together. Running it costs about four cents in AI usage fees per ride and takes about ten minutes, using Google's Gemini AI model by default, though it can also work with OpenAI's models or a model run entirely on your own computer. After processing, a review tool opens in your browser showing short preview clips so you can approve or swap out selections before the final video is put together, or you can skip that step and let the tool decide everything on its own. Setup requires Python, the video tool ffmpeg, and an API key for whichever AI provider you choose. It is aimed at cyclists who record their rides with a GoPro camera and a Garmin bike computer and want an easy way to create a shareable highlight video without spending hours in video editing software.