[{"data":1,"prerenderedAt":487},["ShallowReactive",2],{"project-22154088":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":45,"post_text":47,"what":48,"said":93,"maker_in_thread":233,"shape":244,"grid":248,"attempts":448,"same_week":449,"maker_shows":41,"maker_hidden":479,"reader_version":480,"card":481},"22154088","botium-speech-processing","Text-to-speech and speech-to-text open-source software stack","title","Show HN: Text-to-speech and speech-to-text open-source software stack","https:\u002F\u002Fgithub.com\u002Fcodeforequity-at\u002Fbotium-speech-processing","github.com","ftreml",1580066562,2020,438,85,69,3028,3,"legend","ai","speech recognition and synthesis","api",[24,25,26,27,28],"speech-to-text","text-to-speech","kaldi","marytts","open source speech","curious",0.093,{"positive":32,"neutral":33,"mixed":34,"critical":35},6,13,2,4,25,{"docs":38,"performance":40,"idea":42,"execution":43,"design":44},{"positive":39,"critical":39},0,{"positive":41,"critical":34},1,{"positive":41,"critical":39},{"positive":34,"critical":41},{"positive":39,"critical":41},[18,18,18,39,39,39,41,46,41,18,39,39,39,18,18,18,46,18,39,18,39,18,41,46,39,39,39,39,18,39,39,39,39,39,18,18,46,18,41,46],5,null,{"problem":49,"audience":56,"pricing":47,"motivation":63,"time_to_build":67,"stack":71,"openness":90,"stage":91,"team_size":92},{"value":50,"source":51,"evidence":52},"No easy open-source stack for medium-quality speech recognition and synthesis, e.g. for testing voice IVR systems without cloud services","maker",[53],{"id":54,"by":11,"text":55},"22154090","This project is the result of a one year long learning process in speech recognition and speech synthesis. The original task was to automate the testing of a voice-enabled IVR…",{"value":57,"source":51,"evidence":58},"Developers testing or adding voice to pipelines, especially where cloud services are not allowed",[59,60],{"id":54,"by":11,"text":55},{"id":61,"by":11,"text":62},"22155007","it means: easy to install, easy to use, medium performance. no further know-how needed. compared to the total effort for selecting, training, deploying speech recogniction and…",{"value":64,"source":51,"evidence":65},"Automating tests of a voice-enabled IVR system when cloud solutions were not an option",[66],{"id":54,"by":11,"text":55},{"value":68,"source":51,"evidence":69},"one year of learning process",[70],{"id":54,"by":11,"text":55},[72,76,80,84],{"value":73,"source":51,"evidence":74},"Kaldi",[75],{"id":54,"by":11,"text":55},{"value":77,"source":51,"evidence":78},"MaryTTS",[79],{"id":54,"by":11,"text":55},{"value":81,"source":51,"evidence":82},"Swagger\u002FHTTP JSON API",[83],{"id":54,"by":11,"text":55},{"value":85,"source":51,"evidence":86},"Docker",[87],{"id":88,"by":11,"text":89},"22157811","40GB is maybe too much, but when building the docker images there is some space wasted. The image size after building is around 20GB (12GB marytts, 3GB kaldi de, 6GB kaldi en)","open_source","launched","unknown",{"praise":94,"objections":111,"requests":138,"questions":171,"alternatives":194},[95,103],{"value":96,"source":97,"evidence":98},"Polished demo and well-built work","commenter",[99],{"id":100,"by":101,"text":102},"22156999","tomcam","That's fantastic work and the demo is very well done. Thanks for sharing it. You obviously put a lot of hard work into it. Feels super polished.",{"value":104,"source":51,"evidence":105},"Easy-to-use API that provides a quick voice boilerplate",[106,107],{"id":61,"by":11,"text":62},{"id":108,"by":109,"text":110},"22155805","bArray","> just a small server, hope that it wont crash when posting > the link here In all honesty I was just looking for some pre-processed examples. The API itself seems intuitive…",[112,120,131],{"value":113,"source":97,"evidence":114},"Text-to-speech quality is robotic and weaker than commercial engines",[115,119],{"id":116,"by":117,"text":118},"22157932","ayumu722","I've built a speech synthesis system with Marytts before, it works, but unfortunately, the quality is not very good, HMM and unit selection speech synthesis are now very old…",{"id":108,"by":109,"text":110},{"value":121,"source":97,"evidence":122},"Recognition is less accurate than cloud services and Kaldi is outdated compared to newer models",[123,127],{"id":124,"by":125,"text":126},"22155732","Nimitz14","It is very unlikely a E2E toolkit can outperform kaldi if they're trained on the same data. Also I suspect these guys aren't using the latest kaldi architectures.",{"id":128,"by":129,"text":130},"22155812","z3t4","Cool! Thanks! I've tried both wav2letter and DeepSpeech, and now this, but I get very poor results even with short sentences (compared to Google's proprietary services). Would it…",{"value":132,"source":97,"evidence":133},"Large disk requirement of about 40GB",[134],{"id":135,"by":136,"text":137},"22156573","polishdude20","So why is 40gigs of free space needed?",[139,150,161],{"value":140,"source":97,"evidence":141},"Support for more languages",[142,146],{"id":143,"by":144,"text":145},"22155184","yorwba","You should indicate somewhere which languages are supported. You mention Deepspeech German in the credits, but did you use it to implement support for German or just to inform…",{"id":147,"by":148,"text":149},"22157189","ajaviaad","Which languages are supported?",{"value":151,"source":97,"evidence":152},"Additional STT\u002FTTS backends such as Tacotron2 or wav2letter",[153,157,158],{"id":154,"by":155,"text":156},"22155173","pouta","I built something quite similar on my own product. Is there any interest on adding more STT\u002FTTS backends to the software? Think services like Lyrebird or Trint. I could…",{"id":116,"by":117,"text":118},{"id":159,"by":11,"text":160},"22167798","so this is a real cool project. as soon as you finished your training I will be happy to add it as option (or maybe default setup) in the botium speech processing setup (if you…",{"value":162,"source":97,"evidence":163},"Live demo",[164,168],{"id":165,"by":166,"text":167},"22157808","mariushn","Would love to see a live demo. MaryTTS demo link is broken.",{"id":169,"by":129,"text":170},"22155613","Would be cool with a web demo.",[172,176,187],{"value":173,"source":97,"evidence":174},"Which languages are supported",[175],{"id":147,"by":148,"text":149},{"value":177,"source":97,"evidence":178},"Performance compared to DeepSpeech, wav2letter and cloud services",[179,183],{"id":180,"by":181,"text":182},"22155709","dsteinman","Any idea how it compares to Mozilla's DeepSpeech?",{"id":184,"by":185,"text":186},"22157799","monkeydust","Are there any performance metrics of this versus other offline and cloud based services?",{"value":188,"source":97,"evidence":189},"Real-time streaming support",[190],{"id":191,"by":192,"text":193},"22157346","CommanderData","Any recommendations for a real time solution? I maintain a platform which features live video events we'd like to add captioning and so far can only see IBM Watson providing a…",[195,206,213,219,226],{"value":196,"source":97,"evidence":197},"Mozilla DeepSpeech",[198,202],{"id":199,"by":200,"text":201},"22155659","magicalhippo","> deepspeech was evaluated, but right now kaldi provides better performance Was this before the streaming API was added to DeepSpeech? I recently did some testing with it, and it…",{"id":203,"by":204,"text":205},"22155878","StudentStuff","Mozilla DeepSpeech has seen significant improvements in the quality of results it provides (and RAM\u002FCPU usage) since early December, the accuracy of transcriptions has…",{"value":207,"source":97,"evidence":208},"wav2letter++",[209],{"id":210,"by":211,"text":212},"22155344","grizzles","Facebook has released wav2letter++. I'd wager that will outperform kaldi by a wide margin.",{"value":214,"source":51,"evidence":215},"Google and Amazon cloud speech services",[216],{"id":217,"by":11,"text":218},"22159268","picotts is a command line tool. marytts is a client\u002Fserver tool. (both included in botium speech processing and callable with curl). high quality with google cloud speech and…",{"value":220,"source":97,"evidence":221},"Festival speech synthesis",[222],{"id":223,"by":224,"text":225},"22157368","bobmaxup","If marytts is so good, why are we in many linux distros still using https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FFestival_Speech_Synthesis_System as our default tts system?",{"value":227,"source":97,"evidence":228},"CMU Sphinx",[229],{"id":230,"by":231,"text":232},"22157001","hardwaresofton","Can anyone in the space expand on why it's increasingly rare to see people using\u002Fbuilding on Sphinx[0]? Do people avoid it simply because of an impression that it won't be good…",{"comments":234,"answered":235,"key":236},34,21,[237,238,241],{"id":54,"text":55},{"id":239,"text":240},"22154837","no, it was a special call center software called vacapo.",{"id":242,"text":243},"22155231","good point, will add this information. in short, german and english, because those are the languages i am comfortable with. i hope to find native speakers contributing more…",{"commenters":245,"max_depth":246,"maker_replies":234,"first_comment_mins":39,"lifespan_hours":247},32,10,199,[249,251,255,256,258,260,261,263,266,269,271,274,276,278,280,283,285,288,292,295,299,302,305,308,310,311,313,317,320,323,324,327,329,333,336,339,342,347,348,349,351,353,355,357,358,360,362,365,368,372,375,377,379,381,384,386,387,390,393,395,397,399,402,405,407,409,411,413,415,416,418,421,423,425,427,429,431,434,436,437,439,441,442,444,446],[54,39,39,250,11,41],"project\u002Fneutral\u002Fidea",[252,39,18,253,254,39],"22154591","maker\u002Fneutral\u002Fnone","dillonmckay",[239,39,39,253,11,41],[143,39,18,257,144,39],"project\u002Fneutral\u002Fdocs",[242,39,39,259,11,41],"project\u002Fneutral\u002Fperformance",[199,39,18,259,200,39],[262,39,39,259,11,41],"22155708",[264,39,39,265,200,39],"22155750","commenter\u002Fneutral\u002Fperformance",[267,39,39,268,11,41],"22155799","project\u002Fneutral\u002Fexecution",[203,39,39,270,204,39],"other\u002Fneutral\u002Fperformance",[272,39,39,265,273,39],"22179448","Erlich_Bachman",[275,39,39,270,204,39],"22231687",[116,39,41,277,117,39],"project\u002Fcritical\u002Fperformance",[279,39,39,253,11,41],"22158154",[281,39,46,282,117,39],"22158564","maker\u002Fpositive\u002Fnone",[154,41,46,284,155,39],"project\u002Fpositive\u002Fidea",[286,41,39,287,11,41],"22155211","maker\u002Fpositive\u002Fidea",[289,34,41,290,291,39],"22156767","project\u002Fcritical\u002Fexecution","hajimemash",[293,34,39,294,11,41],"22157803","maker\u002Fneutral\u002Fexecution",[296,34,39,297,298,39],"22160401","other\u002Fneutral\u002Fnone","DonHopkins",[300,18,18,250,301,39],"22157125","sandreas",[303,18,39,304,11,41],"22157291","maker\u002Fneutral\u002Fidea",[306,18,39,307,301,39],"22158135","other\u002Fneutral\u002Fidea",[309,18,39,307,11,41],"22158293",[230,35,39,297,231,39],[312,35,39,304,11,41],"22157046",[314,35,39,315,316,39],"22157070","other\u002Fcritical\u002Fdocs","shakna",[318,35,39,270,319,39],"22157942","sgt101",[321,46,18,259,322,39],"22156167","tianshuo",[191,32,18,253,192,39],[325,32,39,326,11,41],"22157399","project\u002Fpositive\u002Fperformance",[328,32,39,259,11,41],"22157698",[330,331,18,259,332,39],"22157205",7,"briga",[334,331,39,335,11,41],"22157229","project\u002Fmixed\u002Fperformance",[337,331,39,270,338,39],"22165310","lunixbochs",[100,340,46,341,101,39],8,"maker\u002Fpositive\u002Fexecution",[343,344,18,345,346,39],"22154969",9,"maker\u002Fneutral\u002Fdocs","monkpit",[61,344,39,287,11,41],[223,246,39,297,224,39],[350,246,39,277,11,41],"22157388",[135,352,18,253,136,39],11,[354,352,39,259,11,41],"22156977",[356,352,39,265,136,39],"22157529",[88,352,39,268,11,41],[165,359,18,268,166,39],12,[361,359,39,268,11,41],"22157936",[363,359,41,364,166,39],"22158000","project\u002Fcritical\u002Fdesign",[366,359,39,367,11,41],"22158046","maker\u002Fneutral\u002Fdesign",[369,359,46,370,371,39],"22158062","project\u002Fpositive\u002Fexecution","klft",[373,33,39,297,374,39],"22159083","dmos62",[217,33,39,376,11,41],"project\u002Fneutral\u002Fnone",[378,33,39,297,374,39],"22168184",[210,380,39,270,211,39],14,[382,380,39,383,11,41],"22155367","maker\u002Fneutral\u002Fperformance",[385,380,39,265,211,39],"22155428",[180,380,39,270,181,39],[388,380,39,389,11,41],"22155738","other\u002Fneutral\u002Fdocs",[391,380,39,392,200,39],"22155798","commenter\u002Fneutral\u002Fnone",[124,380,39,394,125,39],"commenter\u002Fcritical\u002Fperformance",[396,380,39,335,11,41],"22155789",[398,380,18,383,125,39],"22155917",[400,380,39,401,211,39],"22156587","commenter\u002Fneutral\u002Fidea",[403,380,39,404,11,41],"22157022","commenter\u002Fpositive\u002Fnone",[406,380,39,265,338,39],"22159967",[408,380,39,265,125,39],"22161169",[410,380,39,265,338,39],"22162602",[412,380,39,253,11,41],"22162678",[414,380,39,392,338,39],"22163137",[159,380,39,404,11,41],[417,380,39,265,338,39],"22170173",[419,380,39,420,11,41],"22170455","maker\u002Fmixed\u002Fperformance",[422,380,39,404,125,39],"22163502",[184,424,18,259,185,39],15,[426,424,39,265,11,41],"22157884",[147,428,18,376,148,39],16,[430,428,39,376,11,41],"22157222",[169,432,46,433,129,39],17,"project\u002Fpositive\u002Fnone",[435,432,39,376,11,41],"22155692",[108,432,18,335,109,39],[438,432,18,335,129,39],"22155832",[440,432,39,335,11,41],"22155833",[128,432,41,277,129,39],[443,432,46,326,200,39],"22156325",[445,432,39,392,11,41],"22157016",[447,432,39,259,11,41],"22157003",[],[450,455,460,465,470,475],{"id":451,"name":452,"points":453,"category":454},"22152736","usb-hubs.org",560,"web_saas",{"id":456,"name":457,"points":458,"category":459},"22128385","Women Make",345,"consumer_social",{"id":461,"name":462,"points":463,"category":464},"22164212","Binarysearch.io",279,"learning_science",{"id":466,"name":467,"points":468,"category":469},"22174412","Thinc",250,"languages",{"id":471,"name":472,"points":473,"category":474},"22139975","Postgresqlco.nf",223,"dev_tools",{"id":476,"name":477,"points":478,"category":454},"22145168","Userbase",206,false,"v2",{"id":482,"name":5,"line":6,"ai":479,"cat":20,"year":13,"points":14,"comments":15,"rec":29,"s":483,"top":18,"kw":484,"maker":11,"domain":10,"tier":19,"attempts":39,"strip":485,"url":9,"mix":486,"praise":96,"objection":113,"sub":21},22154088,0.09,[24,25,26,27,28],"3330001513000333530303150000300000335315",[32,33,34,35],1794036585845]