[{"data":1,"prerenderedAt":166},["ShallowReactive",2],{"project-30346348":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":14,"rank_in_year":15,"shown_in_year":16,"top":17,"tier":18,"category":19,"subcategory":20,"format":21,"keywords":22,"reception":27,"sentiment":28,"stance_mix":29,"scored_comments":30,"aspects":32,"strip":35,"post_text":28,"what":37,"said":62,"maker_in_thread":102,"shape":109,"grid":112,"attempts":130,"same_week":131,"maker_shows":127,"maker_hidden":159,"reader_version":160,"card":161},"30346348","Scryer","deep search for video and podcast content","title","Show HN: Scryer - deep search for video and podcast content","https:\u002F\u002Fscryer.xyz","scryer.xyz","prohobo",1644933033,2022,8,2256,2570,88,"shown","data","search engine","web_app",[23,24,25,26],"transcript search","video search","podcast search","entity extraction","quiet",null,{"positive":30,"neutral":31,"mixed":31,"critical":31},4,0,{"idea":33},{"positive":34,"critical":31},3,[36,31,36,31,36,31,31,36],5,{"problem":38,"audience":28,"pricing":28,"motivation":45,"time_to_build":28,"stack":49,"openness":60,"stage":61,"team_size":60},{"value":39,"source":40,"evidence":41},"Multimedia search tools are bad; searching inside video and audio content is hard","maker",[42],{"id":43,"by":11,"text":44},"30346403","Heya, I'm working on creating a metadata search engine (using transcripts) for YouTube, Spotify, etc. Basically any video\u002Faudio content. Everyone always complains about how bad…",{"value":46,"source":40,"evidence":47},"Multimedia search tools are poor and earlier attempts like Blinkx fell short",[48],{"id":43,"by":11,"text":44},[50,56],{"value":51,"source":40,"evidence":52},"Elasticsearch",[53],{"id":54,"by":11,"text":55},"30426134","Hey sorry I didn't see this reply until now! We take the transcripts and run natural language processing models on them to find keywords, n-grams, and pronouns. We then…",{"value":57,"source":40,"evidence":58},"Natural language processing models for keywords, n-grams and entities",[59],{"id":54,"by":11,"text":55},"unknown","beta",{"praise":63,"objections":72,"requests":73,"questions":86,"alternatives":97},[64],{"value":65,"source":66,"evidence":67},"Useful idea; search over YouTube captions across many videos is missing","commenter",[68],{"id":69,"by":70,"text":71},"30351009","HanClinto","This is great! Literally just yesterday I was looking for something like this. I've used the Chrome extension YouTube Captions Search (…",[],[74,80],{"value":75,"source":66,"evidence":76},"Let users contribute caption files to a shared database",[77],{"id":78,"by":70,"text":79},"30351825","Yeah, that makes a lot of sense! Is there any way that we can contribute? Like, if I watch a video with CC turned on, it feels like it would be easy for me to upload the CC…",{"value":81,"source":66,"evidence":82},"Federated network where people link their own feeds",[83],{"id":84,"by":11,"text":85},"30352012","The main bottleneck is the machine learning to extract metadata, the transcriptions are actually handled very fast at an early stage. Someone mentioned that it might be a good…",[87,91],{"value":88,"source":66,"evidence":89},"Where does the transcript data come from",[90],{"id":69,"by":70,"text":71},{"value":92,"source":66,"evidence":93},"What metadata is extracted with machine learning and how large is the model",[94],{"id":95,"by":70,"text":96},"30361752","What sort of metadata are you extracting with machine learning? Doing more than simple keyword indexing? I'll admit, I don't know very much about the particulars of building a…",[98],{"value":99,"source":66,"evidence":100},"YouTube Captions Search Chrome extension",[101],{"id":69,"by":70,"text":71},{"comments":30,"answered":34,"key":103},[104,107,108],{"id":105,"text":106},"30351616","Thanks, actually I'm looking for any feedback I can get on this and then make it as open source as possible (while still earning something) so people can index their own content…",{"id":84,"text":85},{"id":54,"text":55},{"commenters":34,"max_depth":110,"maker_replies":30,"first_comment_mins":30,"lifespan_hours":111},6,165,[113,115,118,119,121,122,123,125],[69,31,36,114,70,31],"project\u002Fpositive\u002Fidea",[105,31,31,116,11,117],"maker\u002Fneutral\u002Fnone",1,[78,31,36,114,70,31],[84,31,31,120,11,117],"project\u002Fneutral\u002Fexecution",[95,31,36,114,70,31],[54,31,31,120,11,117],[43,117,31,124,11,117],"project\u002Fneutral\u002Fidea",[126,127,36,128,129,31],"30352309",2,"project\u002Fpositive\u002Fnone","polyterative",[],[132,136,141,146,150,154],{"id":133,"name":134,"points":135,"category":19},"30290225","EdgeDB 1.0",947,{"id":137,"name":138,"points":139,"category":140},"30332535","iradesign",572,"creative",{"id":142,"name":143,"points":144,"category":145},"30362112","hue.tools",481,"dev_tools",{"id":147,"name":148,"points":149,"category":140},"30356944","Streamline",313,{"id":151,"name":152,"points":153,"category":145},"30364020","ytcast",227,{"id":155,"name":156,"points":157,"category":158},"30349243","Shortwave",224,"productivity",false,"v2",{"id":162,"name":5,"line":6,"ai":159,"cat":19,"year":13,"points":14,"comments":14,"rec":27,"s":28,"top":17,"kw":163,"maker":11,"domain":10,"tier":18,"attempts":31,"strip":164,"url":9,"mix":165,"praise":65,"objection":-1,"sub":20},30346348,[23,24,25,26],"50505005",[30,31,31,31],1794036617009]