[{"data":1,"prerenderedAt":508},["ShallowReactive",2],{"project-40346995":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":27,"sentiment":28,"stance_mix":29,"scored_comments":34,"aspects":35,"strip":44,"post_text":46,"what":47,"said":92,"maker_in_thread":211,"shape":221,"grid":226,"attempts":470,"same_week":471,"maker_shows":41,"maker_hidden":500,"reader_version":501,"card":502},"40346995","Pi-C.A.R.D","A Raspberry Pi Voice Assistant","title","Show HN: Pi-C.A.R.D, a Raspberry Pi Voice Assistant","https:\u002F\u002Fgithub.com\u002Fnkasmanoff\u002Fpi-card","github.com","nkaz123",1715627035,2024,344,92,158,3515,5,"legend","ai","voice assistant","other",[21,24,25,26],"raspberry pi","offline llm","local ai","curious",0.193,{"positive":30,"neutral":31,"mixed":32,"critical":33},8,27,0,2,37,{"privacy":36,"performance":37,"idea":38,"design":40,"execution":42,"docs":43},{"positive":33,"critical":32},{"positive":32,"critical":33},{"positive":39,"critical":32},3,{"positive":41,"critical":32},1,{"positive":32,"critical":32},{"positive":32,"critical":32},[18,32,32,32,32,32,32,32,39,18,18,39,32,32,39,39,45,39,32,32,39,32,39,18,32,39,39,39,39,32,32,39,39,39,32,32,39,39,39,18],4,"Pi-card is an AI powered voice assistant running locally on a Raspberry Pi. It is capable of doing anything a standard LLM (like ChatGPT) can do in a conversational setting. In addition, if there is a camera equipped, you can also ask Pi-card to take a photo, describe what it sees, and then ask questions about that.\n\nIt uses distributed models so latency is something I'm working on, but I am curious on where this could go, if anywhere.\n\nVery much a WIP. Feedback welcome :-)",{"problem":48,"audience":56,"pricing":56,"motivation":57,"time_to_build":56,"stack":61,"openness":90,"stage":91,"team_size":90},{"value":49,"source":50,"evidence":51},"Voice assistant that runs offline without sending user data to third parties","commenter",[52],{"id":53,"by":54,"text":55},"40347538","rkagerer","I wanted to create a voice assistant that is completely offline and doesn't require any internet connection. This is because I wanted to ensure that the user's privacy is…",null,{"value":58,"source":50,"evidence":59},"Wanted a privacy-preserving alternative to cloud voice assistants like Alexa",[60],{"id":53,"by":54,"text":55},[62,68,74,80,86],{"value":63,"source":64,"evidence":65},"Raspberry Pi","maker",[66],{"id":67},"post",{"value":69,"source":64,"evidence":70},"Llama 3 8B",[71],{"id":72,"by":11,"text":73},"40348352","Fully depends on the model, how much conversational context you provide, but if you keep things to a bare minimum, ~\u003C 5 seconds from message received to starting the response…",{"value":75,"source":64,"evidence":76},"Whisper tiny",[77],{"id":78,"by":11,"text":79},"40360925","There are two models at use here, whisper tiny for transcribing audio, and then llama 3 for responding. Whisper tiny is multi lingual (though I am using the english specific…",{"value":81,"source":64,"evidence":82},"espeak",[83],{"id":84,"by":11,"text":85},"40347860","Yes! I'm currently using https:\u002F\u002Fespeak.sourceforge.net\u002F, so it isn't especially fun to listen to though. Additionally, since I'm streaming the LLM response, it won't take long…",{"value":87,"source":64,"evidence":88},"moondream vision model",[89],{"id":72,"by":11,"text":73},"unknown","prototype",{"praise":93,"objections":119,"requests":130,"questions":159,"alternatives":185},[94,109],{"value":95,"source":50,"evidence":96},"Privacy-focused offline alternative to Alexa",[97,101,105],{"id":98,"by":99,"text":100},"40348661","pawelduda","All I need is a voice assistant that: - RPi 4 can handle, - I can integrate with HomeAssistant, - is offline only, and doesn't send my data anywhere. This project seems to be…",{"id":102,"by":103,"text":104},"40351770","8mobile","I really like having a voice assistant that focuses on privacy first. I will definitely try it. always add a video to show how it works. Thank you",{"id":106,"by":107,"text":108},"40350981","ethagnawl","I'm looking forward to trying this. Hopefully this gains traction, as (AFAIK) an open, reliable, flexible, privacy focused voice assistant is still sorely needed. About a year…",{"value":110,"source":50,"evidence":111},"Name and Star Trek theme well received",[112,115],{"id":113,"by":54,"text":114},"40347516","Pi-C.A.R.D is perfect. Read it 100% as Picard, and more recognizable that LCARS.",{"id":116,"by":117,"text":118},"40352613","rob74","Extra kudos for the name - and extra extra for using the good old \"Picard facepalm\" meme. But seriously - the name got my attention, then I read the introduction and thought…",[120,124],{"value":121,"source":64,"evidence":122},"Latency is high, about 5 seconds to start a response and 45 seconds for vision",[123],{"id":72,"by":11,"text":73},{"value":125,"source":64,"evidence":126},"Wake word misdetection, the maker notes mis-heard versions of the wake word",[127],{"id":128,"by":11,"text":129},"40347966","The wake word detection is an interesting problem here. As you can see in the repo, I have a lot of mis-heard versions of the wake word in place, in this case being \"Raspberry\".…",[131,145,152],{"value":132,"source":50,"evidence":133},"Run on a generic Linux box or a laptop or mini PC",[134,138,142],{"id":135,"by":136,"text":137},"40349249","knodi123","Is it possible to run this on a generic linux box? Or if not, are you aware of a similar project that can? I've googled it before, but the space is crowded and the caveats are…",{"id":139,"by":140,"text":141},"40352139","squarefoot","+1. That's the #1 feature I want in any \"assistant\". A question: does it run only on the Pi5 or other (also non Raspberry Pi) boards?",{"id":143,"by":140,"text":144},"40364496","Thanks. My ideal use case would probably involve a mini PC, of which I have a few spares and most of them are faster and can be upgraded to more RAM than the Pi5, which…",{"value":146,"source":50,"evidence":147},"Offload heavy processing to a home server",[148],{"id":149,"by":150,"text":151},"40356476","BizarroLand","I have a decent home server, is there a system like this where the heavy lifting can be run on my home server but the front end piped to a ras-pi with speaker\u002Fcamera, etc. ? If…",{"value":153,"source":50,"evidence":154},"Better text-to-speech than espeak",[155],{"id":156,"by":157,"text":158},"40359738","t-vi","When I did a similar thing (but with less LLM) I liked https:\u002F\u002Fgithub.com\u002Fcoqui-ai\u002FTTS but back then I needed to cut out the conversion step from tensor to a list of numbers to…",[160,164,171,178],{"value":161,"source":50,"evidence":162},"Does it run only on Raspberry Pi 5 or other boards",[163],{"id":139,"by":140,"text":141},{"value":165,"source":50,"evidence":166},"How does the wake word work",[167],{"id":168,"by":169,"text":170},"40348296","aci_12","how does wake word work? Does it keep listening and ignore if the last few seconds does not have the wake word\u002Fphrase?",{"value":172,"source":50,"evidence":173},"Does it support languages other than English",[174],{"id":175,"by":176,"text":177},"40353035","ghnws","I didn't see a mention of languages in the readme. Does this understand languages other than english?",{"value":179,"source":50,"evidence":180},"Can the model learn and keep memory beyond one conversation",[181],{"id":182,"by":183,"text":184},"40351435","zenkalia","How hard is it to make an offline model like this learn? The readme mentions a memory that lasts as long as each conversation which seems like such a hard limitation to live with.",[186,193,200,204],{"value":187,"source":50,"evidence":188},"Home Assistant voice control",[189],{"id":190,"by":191,"text":192},"40354971","jhbruhn","Have you checked out the Voice Assistant functionalities integrated into HA? https:\u002F\u002Fwww.home-assistant.io\u002Fvoice_control\u002F NabuCasa employed the Rhasspy main dev to work on these…",{"value":194,"source":50,"evidence":195},"Rhasspy",[196],{"id":197,"by":198,"text":199},"40349369","baobun","Check out Rhasspy. You won't get anything practically useful running LLMs on the 4B but you also don't strictly need LLM-based models. In the Rhasspy community, a common pattern…",{"value":201,"source":50,"evidence":202},"Mycroft",[203],{"id":106,"by":107,"text":108},{"value":205,"source":50,"evidence":206},"Willow",[207],{"id":208,"by":209,"text":210},"40348239","cmcconomy","Funny, I just picked up a device for use with https:\u002F\u002Fheywillow.io for similar reasons",{"comments":212,"answered":45,"key":213},7,[214,217,218],{"id":215,"text":216},"40360861","I think so. I plan to update the README in the coming days with more info, but realistically this is something you could run on your laptop too, barring some changes to how it…",{"id":128,"text":129},{"id":219,"text":220},"40360899","I'm really inspired reading this and hope it can help! I'm planning to put more work into this. I have a few rough demos of it in action on youtube…",{"commenters":222,"max_depth":223,"maker_replies":212,"first_comment_mins":224,"lifespan_hours":225},64,6,32,116,[227,229,233,237,240,242,245,248,251,254,258,260,262,265,268,270,272,275,278,280,282,284,286,289,293,294,297,298,301,304,306,310,313,316,317,320,321,324,326,329,330,333,336,339,341,344,347,349,352,354,356,359,361,364,367,368,370,371,372,375,379,383,387,389,392,394,395,398,401,404,407,409,412,414,416,421,422,424,426,428,430,432,435,437,440,443,446,448,452,456,461,466],[53,32,18,228,54,32],"maker\u002Fpositive\u002Fprivacy",[230,32,32,231,232,32],"40347917","commenter\u002Fneutral\u002Fnone","ornornor",[234,32,32,235,236,32],"40349350","other\u002Fneutral\u002Fprivacy","pyaamb",[238,32,32,235,239,32],"40349403","herval",[241,32,32,235,236,32],"40349490",[243,32,32,235,244,32],"40349764","sneak",[246,32,32,235,247,32],"40350787","Rebelgecko",[249,32,32,235,250,32],"40351019","JakeStone",[252,32,32,235,253,32],"40363314","j5155",[255,32,32,256,257,32],"40349475","other\u002Fneutral\u002Fnone","abraae",[259,32,32,256,236,32],"40349513",[261,32,32,256,257,32],"40350105",[263,32,32,235,264,32],"40350325","krono",[266,32,32,235,267,32],"40355711","jerbear4328",[269,32,32,235,264,32],"40358651",[271,32,32,235,253,32],"40363323",[273,32,32,235,274,32],"40399571","mananaysiempre",[276,32,32,235,277,32],"40356464","everforward",[139,32,39,279,140,32],"project\u002Fneutral\u002Fnone",[215,32,32,281,11,41],"maker\u002Fneutral\u002Fnone",[143,32,39,283,140,32],"project\u002Fneutral\u002Fperformance",[116,32,18,285,117,32],"project\u002Fpositive\u002Fidea",[287,32,32,256,288,32],"40353050","gpderetta",[290,32,39,291,292,32],"40360971","project\u002Fneutral\u002Fdesign","fat_cantor",[98,41,18,285,99,32],[295,41,39,279,296,32],"40348762","8xeh",[197,41,39,283,198,32],[299,41,32,256,300,32],"40352286","shaan7",[302,41,32,256,303,32],"40361173","nirav72",[305,41,32,256,300,32],"40364708",[307,41,32,308,309,32],"40355120","other\u002Fneutral\u002Fperformance","phkahler",[311,41,32,256,312,32],"40355448","genewitch",[314,41,32,315,277,32],"40356625","commenter\u002Fneutral\u002Fperformance",[149,41,39,291,150,32],[318,41,39,279,319,32],"40349521","yangikan",[190,41,39,279,191,32],[322,33,39,279,323,32],"40347498","eddieroger",[113,33,18,325,54,32],"project\u002Fpositive\u002Fdesign",[327,33,39,279,328,32],"40347783","orthecreedence",[128,33,32,281,11,41],[331,33,39,332,191,32],"40354981","project\u002Fneutral\u002Fexecution",[334,33,32,256,335,32],"40348695","thesnide",[337,33,32,256,338,32],"40349531","pimeys",[340,33,32,256,335,32],"40358477",[342,33,32,256,343,32],"40350855","fnordpiglet",[345,33,32,256,346,32],"40353125","TeMPOraL",[348,33,32,231,343,32],"40356823",[350,33,39,279,351,32],"40351326","datadrivenangel",[353,33,32,231,117,32],"40352639",[355,33,32,231,323,32],"40355620",[357,33,39,279,358,32],"40347807","bdcravens",[360,33,32,235,346,32],"40353144",[362,33,32,256,363,32],"40347828","MisterTea",[365,33,39,279,366,32],"40347922","layer8",[106,39,18,285,107,32],[369,39,32,256,312,32],"40355531",[219,39,32,281,11,41],[135,45,39,332,136,32],[373,45,39,332,374,32],"40354508","CaptainOfCoit",[376,18,18,377,378,32],"40349587","project\u002Fpositive\u002Fnone","MH15",[380,223,39,381,382,32],"40352678","project\u002Fneutral\u002Fidea","robbyiq999",[384,223,41,385,386,32],"40353544","project\u002Fcritical\u002Fperformance","piltdownman",[388,223,32,256,312,32],"40355497",[390,212,39,291,391,32],"40347379","harwoodr",[84,212,32,393,11,41],"maker\u002Fneutral\u002Fexecution",[156,212,39,393,157,32],[396,30,32,256,397,32],"40349629","stereosteve",[399,30,32,256,400,32],"40349673","dragonwriter",[402,30,32,256,403,32],"40350332","lttlrck",[405,30,32,256,406,32],"40351005","pkaye",[408,30,32,256,346,32],"40353168",[410,30,32,256,411,32],"40355205","cheeseomlit",[168,413,39,291,169,32],9,[415,413,39,332,136,32],"40349231",[417,418,39,419,420,32],"40348329",10,"maker\u002Fneutral\u002Fperformance","dasl",[72,418,32,283,11,41],[182,423,39,381,183,32],11,[208,425,32,256,209,32],12,[427,425,32,256,136,32],"40349237",[429,425,32,256,11,41],"40349972",[431,425,32,256,136,32],"40351528",[433,425,32,256,434,32],"40354090","kkielhofner",[436,425,41,385,434,32],"40354319",[102,438,18,439,103,32],13,"project\u002Fpositive\u002Fprivacy",[175,441,39,442,176,32],14,"project\u002Fneutral\u002Fdocs",[444,441,39,442,445,32],"40353253","timendum",[78,441,32,447,11,41],"maker\u002Fneutral\u002Fdocs",[449,450,32,256,451,32],"40375617",15,"alexalx666",[453,454,39,332,455,32],"40352431",16,"nl",[457,458,39,459,460,32],"40350593",17,"maker\u002Fneutral\u002Fidea","kazinator",[462,463,18,464,465,32],"40349984",18,"maker\u002Fpositive\u002Fnone","ddingus",[467,468,32,256,469,32],"40349776",19,"nickthegreek",[],[472,477,482,487,492,496],{"id":473,"name":474,"points":475,"category":476},"40318542","Jam",886,"dev_tools",{"id":478,"name":479,"points":480,"category":481},"40331968","Omniclip",588,"creative",{"id":483,"name":484,"points":485,"category":486},"40307519","Hackerverse",520,"data",{"id":488,"name":489,"points":490,"category":491},"40338275","Cap",351,"productivity",{"id":493,"name":494,"points":495,"category":20},"40345696","Pipecat",346,{"id":497,"name":498,"points":499,"category":486},"40367090","Quary",293,false,"v2",{"id":503,"name":5,"line":6,"ai":500,"cat":20,"year":13,"points":14,"comments":15,"rec":27,"s":504,"top":18,"kw":505,"maker":11,"domain":10,"tier":19,"attempts":32,"strip":506,"url":9,"mix":507,"praise":95,"objection":121,"sub":21},40346995,0.19,[21,24,25,26],"5000000035530033430030350333300333003335",[30,31,32,33],1793991091439]