{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/8cee810a6c3a4111aadfb423077498a4\" frameborder=\"0\" width=\"1658\" height=\"1243\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1243,"width":1658,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1243,"thumbnail_width":1658,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/8cee810a6c3a4111aadfb423077498a4-14fa6302c5c11bca.gif","duration":293.59,"title":"Tuneweave Music recommender","description":"In this Loom, I introduce Tuneweeb, a music recommender that takes natural language requests and returns five grounded song recommendations from a multilingual catalog. I explain the pipeline, an LLM parses my text, a Sentence Transformer retrieves 40 similar songs, a Rhythmistic Ranker scores audio features to pick the best five, and an LLM generates explanations strictly from catalogue descriptions. I also show refinements like Not Lana Del Rey, add English, and avoid a specific track. There was no direct viewer action requested, aside from trying those refinement commands."}