video1.0<iframe src="https://www.loom.com/embed/eae8e365e71a489e872152db640f4e63" frameborder="0" width="1920" height="1440" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14401920Loomhttps://www.loom.com14401920https://cdn.loom.com/sessions/thumbnails/eae8e365e71a489e872152db640f4e63-b78b5717f939c4cf.gif617.621In this video, I introduce a grounded voice wine advisor that I developed using a wine dataset. The system retrieves the right wines based on user preferences, budget, and occasion, delivering recommendations in both text and natural voice. I showcase a live demo where I provide options for various price ranges and regions, highlighting wines like the Chateau Lafite Rothschild 2014 and the Barefoot Cellars Cabernet Sauvignon. I explain the technical challenges I faced, including ensuring the model generates grounded answers from the actual wine catalog. I invite viewers to explore the demo and provide feedback on their experience.