[{"data":1,"prerenderedAt":379},["ShallowReactive",2],{"project-38705535":3},{"id":4,"name":5,"line":6,"line_source":7,"title":8,"url":9,"domain":10,"maker":11,"posted_at":12,"year":13,"points":14,"comments":15,"rank_in_year":16,"shown_in_year":17,"top":18,"tier":19,"category":20,"subcategory":21,"format":22,"keywords":23,"reception":29,"sentiment":30,"stance_mix":31,"scored_comments":36,"aspects":37,"strip":48,"post_text":50,"what":51,"said":102,"maker_in_thread":207,"shape":218,"grid":222,"attempts":340,"same_week":341,"maker_shows":49,"maker_hidden":371,"reader_version":372,"card":373},"38705535","AI Video Search Engine (avse)","An open source AI video search engine to learn more about AI","title","Show HN: I built an open source AI video search engine to learn more about AI","https:\u002F\u002Favse.vercel.app","avse.vercel.app","yoeven",1703047494,2023,145,49,434,3109,14,"notable","ai","semantic video search","web_app",[24,25,26,27,28],"video search","semantic search","rag","embeddings","transcript search","debated",0.016,{"positive":32,"neutral":33,"mixed":34,"critical":35},7,9,0,6,22,{"idea":38,"performance":41,"pricing":43,"execution":44,"design":45,"docs":47},{"positive":39,"critical":40},4,2,{"positive":40,"critical":42},3,{"positive":34,"critical":34},{"positive":34,"critical":34},{"positive":46,"critical":34},1,{"positive":34,"critical":46},[46,34,34,49,49,46,34,46,42,34,34,34,34,42,34,49,49,49,34,34,42,42,34,46,42,34,34,34,42,42,34,46,49,34,42,49,34,42,34,46],5,"Hi there! When Supabase announced their recent hackathon, I thought it was a good time to build something to learn more about so many of the new AI models and tech out there. From the different techniques of embedding documents to the future RAG.\n\nWith the rise of short form content with TikTok and Youtube. A lot more knowledge is in videos than ever before. Finding specific answers within millions of videos can be difficult for any one person to go through. So the question is if there is Google that indexes text on website making it easier to find based on the context of on your question, why",{"problem":52,"audience":58,"pricing":67,"motivation":73,"time_to_build":77,"stack":78,"openness":99,"stage":100,"team_size":101},{"value":53,"source":54,"evidence":55},"Finding specific answers buried inside millions of videos is hard","maker",[56],{"id":57},"post",{"value":59,"source":60,"evidence":61},"People searching for specific knowledge within videos","inferred",[62,63],{"id":57},{"id":64,"by":65,"text":66},"38706494","remuskaos","This is great, I've been looking for exactly that! Often times there are \"hidden gems\" of knowledge in longer videos, for example how some calculations for a specific problem are…",{"value":68,"source":54,"evidence":69},"Free to index videos after a quick email sign-up",[70],{"id":71,"by":11,"text":72},"38706874","Hey there! Thanks a lot :) I've gone ahead and indexed the video you shared and you can try searching something like: \"how does heat sink thermal calculation work\" as it tends to…",{"value":74,"source":54,"evidence":75},"Learn about new AI models, embeddings and RAG",[76],{"id":57},null,[79,83,87,91,95],{"value":80,"source":54,"evidence":81},"Supabase (PostgreSQL, pgvector, Auth)",[82],{"id":57},{"value":84,"source":54,"evidence":85},"Hasura (GraphQL layer)",[86],{"id":57},{"value":88,"source":54,"evidence":89},"Fly (hosts Hasura)",[90],{"id":57},{"value":92,"source":54,"evidence":93},"JigsawStack (summary and chat AI)",[94],{"id":57},{"value":96,"source":54,"evidence":97},"Vercel (Next.js hosting, serverless functions)",[98],{"id":57},"open_source","launched","unknown",{"praise":103,"objections":115,"requests":145,"questions":169,"alternatives":185},[104,111],{"value":105,"source":106,"evidence":107},"Finds relevant sections of long videos well","commenter",[108],{"id":109,"by":65,"text":110},"38707166","Thank you for answering and indexing the video so quickly. I'm blown away how good this works! I searched for the query just like you said and it really found exactly the…",{"value":112,"source":106,"evidence":113},"Solves the hidden-gems problem in long videos",[114],{"id":64,"by":65,"text":66},[116,126,134],{"value":117,"source":106,"evidence":118},"Ranking and relevance are weak; search is not a solved problem",[119,123],{"id":120,"by":121,"text":122},"38710644","ZephyrBlu","The problem with this project is that it doesn't really solve the valuable problem, which is ranking\u002Frelevance. It's effectively just a very simple \"indexer\", if you can even…",{"id":124,"by":121,"text":125},"38718861","> but wouldn't able to accurately contextualise the actual content of the video based on the users question which is the focus on the tool Are you sure this is what users want?…",{"value":127,"source":106,"evidence":128},"Results are sensitive to exact query phrasing",[129,130],{"id":109,"by":65,"text":110},{"id":131,"by":132,"text":133},"38707891","swores","It doesn't sound ideal that it's so phrasing dependant - 5th on a list of 9 videos is poor if you're searching without knowing which video you're looking for and without knowing…",{"value":135,"source":106,"evidence":136},"Only indexes selected channels, so many queries return irrelevant results",[137,141],{"id":138,"by":139,"text":140},"38710364","eleijonmarck","It does not work at indexing anything else than AI? Tried searching for something like zero knowledge proofs and it showed me courses in Java. Interesting approach though",{"id":142,"by":143,"text":144},"38710820","mvdtnz","I haven't been able to get a good search result yet. I tried \"how do i improve my lap times at bathurst in assetto corsa competizione\" and got a couple of irrelevant results…",[146,150,154,165],{"value":147,"source":54,"evidence":148},"Support for TikTok and automatic indexing",[149],{"id":57},{"value":151,"source":106,"evidence":152},"Include or exclude channels in results",[153],{"id":109,"by":65,"text":110},{"value":155,"source":106,"evidence":156},"Dark mode and topic tabs in the UI",[157,161],{"id":158,"by":159,"text":160},"38709675","altruios","This is wonderful. A fantastic step towards indexing video... I have some ideas I'd like to share - something to improve this project: 1: Are you only scraping the video id and…",{"id":162,"by":163,"text":164},"38711301","samstave","This is amazing... UX Feedback: The vid bubbles are HUGE. https:\u002F\u002Fi.imgur.com\u002FTjHhtRr.png It would be great to have TopicTabs... I search for \"X\" and it gives me one of the topic…",{"value":166,"source":54,"evidence":167},"Transcribe audio with Whisper for videos without captions",[168],{"id":57},[170,177],{"value":171,"source":106,"evidence":172},"How were chunk sizes chosen for RAG",[173],{"id":174,"by":175,"text":176},"38707769","social_quotient","How did you decide the chunking size? I’m working on a similar project and it seems our sweet spot was around 800 . But would really love to hear what others are doing here for…",{"value":178,"source":106,"evidence":179},"How to add videos to the index",[180,181],{"id":64,"by":65,"text":66},{"id":182,"by":183,"text":184},"38711892","thisisit","I like this. Another idea maybe is to build a podcast search engine too. The descriptions are there but many times the content is built up to ensure a particular podcast running…",[186,193,200],{"value":187,"source":106,"evidence":188},"askYouTube.ai",[189],{"id":190,"by":191,"text":192},"38708475","twayt","Also checkout www.askYouTube.ai Which does essentially this but requires no indexing of videos!",{"value":194,"source":106,"evidence":195},"dexa.ai",[196],{"id":197,"by":198,"text":199},"38717511","umate","Have you tried dexa.ai? Here is an example with Hubermanlab Podcast - https:\u002F\u002Fdexa.ai\u002Fhuberman",{"value":201,"source":106,"evidence":202},"videogist",[203],{"id":204,"by":205,"text":206},"38708289","viddi","There is a service that produces descriptions of each scene: https:\u002F\u002Fwww.videogist.co\u002F But I don't know about searchability.",{"comments":208,"answered":209,"key":210},16,8,[211,214,215],{"id":212,"text":213},"38716591","Yeah I agree to an extent. Using a traditional search engine would be simpler and easier to implement but wouldn't able to accurately contextualise the actual content of the…",{"id":71,"text":72},{"id":216,"text":217},"38709455","Technically the other videos do talk about the above query, just that not enough videos of it have been indexed to produce a better result. While yes promoting\u002Fquerying tends…",{"commenters":219,"max_depth":49,"maker_replies":208,"first_comment_mins":220,"lifespan_hours":221},25,150,125,[223,225,227,229,231,233,235,237,239,242,245,249,253,256,258,260,262,264,265,268,269,272,274,276,279,280,282,283,286,287,291,292,294,296,299,303,307,309,311,313,315,317,321,323,325,327,331,333,338],[120,34,46,224,121,34],"project\u002Fcritical\u002Fidea",[212,34,34,226,11,46],"project\u002Fneutral\u002Fidea",[124,34,34,228,121,34],"commenter\u002Fneutral\u002Fidea",[64,46,49,230,65,34],"project\u002Fpositive\u002Fidea",[71,46,34,232,11,46],"maker\u002Fneutral\u002Fdocs",[109,46,49,234,65,34],"project\u002Fpositive\u002Fperformance",[131,46,46,236,132,34],"project\u002Fcritical\u002Fperformance",[216,46,34,238,11,46],"project\u002Fneutral\u002Fperformance",[240,40,46,224,241,34],"38711317","jstartup",[243,40,34,244,11,46],"38716560","maker\u002Fneutral\u002Fnone",[246,42,42,247,248,34],"38706315","project\u002Fneutral\u002Fpricing","Racing0461",[250,42,34,251,252,34],"38706688","other\u002Fneutral\u002Fnone","3abiton",[254,42,34,251,255,34],"38706934","eurekin",[257,42,34,251,252,34],"38712641",[259,42,34,251,255,34],"38720406",[261,42,34,238,11,46],"38707020",[174,42,42,263,175,34],"maker\u002Fneutral\u002Fexecution",[190,39,34,251,191,34],[266,39,49,234,267,34],"38710535","maroonblazer",[158,49,49,230,159,34],[270,49,34,271,11,46],"38710513","maker\u002Fneutral\u002Fidea",[162,35,49,273,163,34],"project\u002Fpositive\u002Fdesign",[275,35,34,244,11,46],"38716614",[277,32,34,251,278,34],"38717578","6510",[182,209,42,226,183,34],[281,209,42,244,241,34],"38711976",[197,209,34,251,198,34],[284,209,34,285,241,34],"38718149","other\u002Fcritical\u002Fperformance",[142,33,46,236,143,34],[288,289,42,226,290,34],"38708146",10,"em-bee",[204,289,34,251,205,34],[293,289,34,271,11,46],"38708993",[295,289,34,251,290,34],"38715926",[297,289,34,298,11,46],"38761122","maker\u002Fpositive\u002Fnone",[300,301,42,226,302,34],"38710720",11,"threeone97",[304,305,42,244,306,34],"38706208",12,"knrz",[308,305,34,244,11,46],"38706983",[138,310,46,236,139,34],13,[312,310,34,244,11,46],"38710573",[314,18,49,230,252,34],"38706677",[316,18,34,244,11,46],"38706997",[318,319,42,226,320,34],"38708977",15,"prakhar897",[322,319,34,244,11,46],"38709387",[324,208,49,230,255,34],"38706613",[326,208,34,244,11,46],"38709407",[328,329,42,226,330,34],"38706540",17,"maCDzP",[332,329,34,271,11,46],"38707032",[334,335,46,336,337,34],"38706941",18,"project\u002Fcritical\u002Fdocs","politelemon",[339,335,34,232,11,46],"38706990",[],[342,347,352,357,361,366],{"id":343,"name":344,"points":345,"category":346},"38706299","moddio2",318,"games",{"id":348,"name":349,"points":350,"category":351},"38647938","Airdraw",269,"creative",{"id":353,"name":354,"points":355,"category":356},"38650570","go-sqlite3",234,"languages",{"id":358,"name":359,"points":360,"category":20},"38679453","Microagents",221,{"id":362,"name":363,"points":364,"category":365},"38673973","Google Earth as glTF",177,"dev_tools",{"id":367,"name":368,"points":369,"category":370},"38669781","NowDo",103,"productivity",false,"v2",{"id":374,"name":5,"line":6,"ai":371,"cat":20,"year":13,"points":14,"comments":15,"rec":29,"s":375,"top":18,"kw":376,"maker":11,"domain":10,"tier":19,"attempts":34,"strip":377,"url":9,"mix":378,"praise":105,"objection":117,"sub":21},38705535,0.02,[24,25,26,27,28],"1005510130000305550033013000330150350301",[32,33,34,35],1793991080455]