{"id":"https://openalex.org/W7225753087","doi":"https://doi.org/10.5281/zenodo.24109988","title":"Hebrew printed-text recognition for Tesseract, fine-tuned from tessdata_best heb on real public-domain NLI scans with Project Ben-Yehuda text (HeOCR R6)","display_name":"Hebrew printed-text recognition for Tesseract, fine-tuned from tessdata_best heb on real public-domain NLI scans with Project Ben-Yehuda text (HeOCR R6)","publication_year":2026,"publication_date":"2026-11-10","ids":{"openalex":"https://openalex.org/W7225753087","doi":"https://doi.org/10.5281/zenodo.24109988"},"language":"en","primary_location":{"id":"doi:10.5281/zenodo.24109988","doi":"https://doi.org/10.5281/zenodo.24109988","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24109988","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"volume":null,"type":"article","is_traditional":true,"indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.5281/zenodo.24109988","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5161045255","display_name":"Shay Palachy Affek","orcid":"https://orcid.org/0000-0002-1459-9320","observed_orcids":["https://orcid.org/0000-0002-1459-9320"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Palachy Affek, Shay","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0002-1459-9320","affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5161072239","display_name":"HeOCR maintainers","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"HeOCR maintainers","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.0,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10601","display_name":"Handwritten Text Recognition Techniques","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10601","display_name":"Handwritten Text Recognition Techniques","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12707","display_name":"Vehicle License Plate Recognition","score":9.999999747378752e-05,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.0,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[],"concepts":[{"id":"https://openalex.org/C91304198","wikidata":"https://www.wikidata.org/wiki/Q9288","display_name":"Hebrew","level":2,"score":0.858299970626831},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6984000205993652},{"id":"https://openalex.org/C43364308","wikidata":"https://www.wikidata.org/wiki/Q8799","display_name":"Byte","level":2,"score":0.5713000297546387},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4618000090122223},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3831999897956848},{"id":"https://openalex.org/C546480517","wikidata":"https://www.wikidata.org/wiki/Q167555","display_name":"Optical character recognition","level":3,"score":0.3483000099658966},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3352999985218048},{"id":"https://openalex.org/C199519371","wikidata":"https://www.wikidata.org/wiki/Q942695","display_name":"Source lines of code","level":3,"score":0.32190001010894775},{"id":"https://openalex.org/C198352243","wikidata":"https://www.wikidata.org/wiki/Q37105","display_name":"Line (geometry)","level":2,"score":0.3206000030040741},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.30720001459121704},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.3021000027656555},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.29829999804496765},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.2930000126361847},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.2881999909877777},{"id":"https://openalex.org/C179926584","wikidata":"https://www.wikidata.org/wiki/Q207714","display_name":"Transcription (linguistics)","level":2,"score":0.28450000286102295},{"id":"https://openalex.org/C112876837","wikidata":"https://www.wikidata.org/wiki/Q837518","display_name":"Alphabet","level":2,"score":0.2736000120639801},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.26159998774528503},{"id":"https://openalex.org/C103088060","wikidata":"https://www.wikidata.org/wiki/Q1062839","display_name":"Error detection and correction","level":2,"score":0.2513999938964844}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.5281/zenodo.24109988","doi":"https://doi.org/10.5281/zenodo.24109988","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24109988","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},{"id":"doi:10.5281/zenodo.24109989","doi":"https://doi.org/10.5281/zenodo.24109989","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24109989","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.5281/zenodo.24109988","doi":"https://doi.org/10.5281/zenodo.24109988","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24109988","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"HeOCR":[0,18,182,188],"R6:":[1],"Hebrew":[2,79,204,232,246,288,1115,1119],"printed-text":[3],"recognition":[4,143,228],"for":[5,14,77,235,245,520,804,1089,1389],"Tesseract":[6,10,76,147,153,168,336,479,719,1004,1181,1216,1249,1437,1458],"R6":[7,255,275,426,1145,1231,1236,1279,1331,1390,1399],"is":[8,53,69,117,303,499,735,918,948,967,1055,1097,1312,1327,1332,1350,1376,1378,1387,1421,1461,1466],"a":[9,70,82,346,410,713,730,805,1130],"language":[11,356,497],"model":[12,26,171,334],"(.traineddata)":[13],"printed":[15,78,88,203,231,806,843,1114],"modern":[16],"Hebrew.":[17,319],"made":[19],"it":[20,93,111,221,1335,1373],"by":[21,181,239,685,1282,1286],"fine-tuning":[22],"Tesseract's":[23,225,367,452,707,956,970,1325],"own":[24,489,951,1356],"heb":[25,423,909,1217,1224],"from":[27,37,100,469,553,566,654,798,941,959,1117,1346,1434],"tessdata_best":[28,172,908,1069,1223,1262],"on":[29,74,119,248,258,277,300,403,430,681,1006,1026,1042,1148,1305,1363,1468],"13,648":[30,535],"real":[31,87,249,301,536,1113],"scanned":[32,683],"lines.":[33,327],"The":[34,266,369,496,756,840,916,937,946,1074,1309],"lines":[35,237,434,564,723,862,867,1068,1344],"come":[36,565,1345],"public-domain":[38,669],"books":[39,206,777,832,842],"in":[40,209,222,309,324,345,387,486,516,559,811,834,838,969,1021],"the":[41,56,61,95,105,149,210,312,325,333,355,413,474,500,517,539,563,628,631,666,672,696,710,722,799,812,816,831,835,849,919,926,929,933,949,974,1002,1017,1022,1027,1035,1038,1043,1066,1101,1276,1306,1313,1351,1358,1364,1368,1394,1406,1422,1426,1448,1451,1469],"National":[42,632],"Library":[43,633],"of":[44,60,108,202,224,229,252,412,432,435,477,542,556,562,630,634,671,709,715,767,820,868,1010,1051,1159,1320,1341,1415,1425],"Israel's":[45],"\"public":[46,646,872],"domain,":[47,647,873],"free":[48,648,874],"use\"":[49,649,661,875],"collection.":[50],"Each":[51,640,987],"line":[52,233,392,404,537,545,684,711,976,988],"labelled":[54],"with":[55,65,90,122,167,332,341,421,425,513,706,788,914,955,993,1037,1122,1180,1334,1442,1447],"Project":[57,667],"Ben-Yehuda":[58,668],"transcription":[59,670],"same":[62,673,1067,1395,1470],"edition,":[63],"never":[64],"OCR":[66,201,247,736],"output.":[67,313,737],"It":[68,116,511,1096,1293],"small,":[71],"measured":[72],"improvement":[73],"stock":[75,422,1190,1248,1291],"pages,":[80,205,1193,1278,1296,1366],"not":[81,281,317,399,528,979,1024,1098,1104,1380,1446],"new":[83],"engine.":[84],"On":[85,104,406,1065,1275],"38":[86,1112,1204,1207,1239,1277,1365],"pages":[89,110,302,1108,1116,1179,1205,1208,1211,1214,1240,1245,1314],"human-adjudicated":[91,1123],"references":[92,1124,1407],"lowers":[94,1280],"character":[96,308],"error":[97,1326,1386],"rate":[98,1019],"(CER)":[99],"0.226":[101,1219],"to":[102,125,329,943,985,991,1085],"0.209.":[103],"10":[106,1210,1213,1244],"hardest":[107,1307],"those":[109],"gains":[112],"nothing":[113],"(see":[114,190,271],"Evaluation).":[115],"released":[118],"its":[120,703,789,911,1355,1397,1416],"own,":[121,921],"no":[123,161,733,760,763,857,1137],"link":[124],"HeOCR's":[126,543,686,1321],"kraken":[127],"pipeline":[128],"(R4d,":[129],"A1,":[130],"R3h).":[131],"File":[132],"heocr_r6.traineddata":[133,344,485,1088],"(trained":[134],"as":[135,214,748,751,1396],"heb.traineddata),":[136],"3,628,595":[137],"bytes":[138,1102],"sha256":[139,177],"73eb18e03f8793dff9c03699db0845779f6b5db4897824bcd61787245ec66e10":[140],"Model":[141],"type":[142],"(line-level":[144],"LSTM,":[145],"CTC);":[146],"does":[148],"page":[150,361,370,1359],"layout":[151],"Format":[152],"4/5":[154],"LSTM":[155,157,912,971],"traineddata,":[156],"only":[158,694,720],"(--oem":[159],"1;":[160],"legacy":[162],"engine).":[163],"Trained":[164],"and":[165,192,207,264,285,290,350,354,424,492,702,724,762,890,905,932,1100,1153,1163,1284,1300,1330,1438],"tested":[166],"5.5.3":[169],"Base":[170],"heb.traineddata":[173],"at":[174,699,712,716,729,818,1057],"e12c65a":[175],"(Apache-2.0),":[176],"dbaa827aea6bc21215638447f17783a1004987c2d0bf5573d111fee397abdae5,":[178],"fine-tuned":[179],"(modified)":[180],"Licence":[183],"Apache-2.0,":[184],"Copyright":[185],"(c)":[186],"2026":[187],"maintainers":[189],"LICENSE":[191],"NOTICE)":[193],"Training":[194,272,530,532,903],"record":[195,569],"r6_training_run.json":[196],"(heocr_tesseract_training_run.v1)":[197],"Code":[198],"HeOCR/heocr-engine":[199],"Uses":[200],"periodicals":[208],"square":[211],"script,":[212],"such":[213],"early-20th-century":[215],"book":[216],"type,":[217],"without":[218,503],"niqqud.":[219],"Use":[220],"place":[223],"heb.":[226],"Line":[227],"single":[230],"images,":[234,538],"example":[236,521,802],"found":[238],"another":[240,1439],"segmenter.":[241],"A":[242],"research":[243],"baseline":[244],"scans.":[250],"Out":[251],"scope":[253],"Handwriting.":[254],"was":[256,280,323,527,663,989,1146],"trained":[257,864],"print":[259,800],"only.":[260],"Niqqud":[261,744],"(vowel":[262],"points)":[263],"cantillation.":[265],"training":[267,326,757,972,1094,1343],"labels":[268,758],"have":[269],"none":[270,1319],"data).":[273],"How":[274,328],"behaves":[276],"pointed":[278,776],"text":[279,322,460,704,795],"measured.":[282],"Rashi":[283],"script":[284],"other":[286,293,457,514,1444],"non-square":[287],"scripts,":[289],"medieval":[291],"or":[292,338,472,483,1140],"historical":[294],"hands.":[295],"Unsupervised":[296],"transcription.":[297],"Its":[298],"CER":[299,1162,1281],"still":[304],"about":[305,1393,1401,1409],"0.21":[306],"(one":[307],"five).":[310],"Check":[311],"Documents":[314],"that":[315,352,408],"are":[316,810,833,846,1103,1129,1156,1413],"mainly":[318],"No":[320,658,1303],"Latin":[321,764],"get":[330],"started":[331],"Install":[335],"4":[337],"5":[339],"(tested":[340],"5.5.3).":[342],"Put":[343],"directory,":[347],"say":[348],"DIR,":[349,482],"pass":[351],"directory":[353,491],"name":[357,498,502],"heocr_r6.":[358],"Whole":[359],"pages:":[360],"segmentation":[362],"mode":[363],"3":[364,383],"(automatic":[365],"layout,":[366],"default).":[368],"figures":[371],"below":[372],"use":[373,400],"it.":[374],"tesseract":[375,439,994],"page.png":[376],"out":[377,441,878,1158],"-l":[378,442,522],"heocr_r6":[379,443,505],"--tessdata-dir":[380,444,450],"DIR":[381,445],"--psm":[382,446,996],"This":[384],"writes":[385],"out.txt,":[386],"logical":[388,742,975],"(reading)":[389],"order.":[390,743,1382],"Single":[391],"images:":[393],"psm":[394,401,416,1183],"13":[395,447,997],"(raw":[396],"line).":[397],"Do":[398],"7":[402,417],"crops.":[405],"crops":[407],"carry":[409],"sliver":[411],"neighbouring":[414],"line,":[415,787],"returns":[418],"empty":[419],"text,":[420],"alike:":[427],"this":[428],"happened":[429],"473":[431],"4,046":[433,901],"one":[436,785,854],"held-out":[437],"book.":[438],"line.png":[440],"Notes:":[448],"-":[449,495],"replaces":[451],"data":[453,533],"directory.":[454],"Output":[455],"formats":[456],"than":[458,826,1405],"plain":[459],"(tsv,":[461],"hocr,":[462],"pdf,":[463],"alto)":[464],"read":[465],"their":[466,1154],"config":[467],"files":[468,958],"DIR/configs.":[470],"Link":[471],"copy":[473],"configs":[475],"folder":[476],"your":[478,487],"installation":[480],"into":[481],"put":[484],"installation's":[488],"tessdata":[490],"drop":[493],"--tessdata-dir.":[494],"file":[501],".traineddata.":[504],"follows":[506],"tessdata's":[507],"naming":[508],"(chi_sim,":[509],"deu_latf).":[510],"combines":[512],"models":[515],"usual":[518],"way,":[519],"heocr_r6+eng,":[523],"which":[524,560,1349],"runs":[525,982],"but":[526,1367,1445],"evaluated.":[529],"details":[531],"Lines:":[534],"train":[540],"split":[541],"bynli_lines_v1":[544],"set":[546,1025,1132],"(Ben-Yehuda":[547],"\u00d7":[548],"NLI).":[549],"They":[550,651,1128],"were":[551,652,692,780],"cut":[552],"NLI":[554,568,871],"scans":[555],"five":[557],"books,":[558,876,1348],"99.9%":[561],"three:":[567],"Book":[570],"Lines":[571,691],"990020308000205171":[572],"Yeshayahu":[573],"Berashadsky":[574],"(1871\u20131908),":[575],"Ketavim":[576],"Aharonim,":[577],"Warsaw:":[578,612],"Tushiya,":[579,898],"1910":[580],"6,076":[581],"990020495520205171":[582],"Shmuel":[583,595],"Leib":[584],"Citron":[585],"(1860\u20131930),":[586],"Me-Ahorei":[587],"ha-Pargod,":[588],"Vilna:":[589],"Z.":[590],"Matz,":[591],"1923\u20131925":[592],"6,052":[593],"990020522850205171":[594],"Tchernowitz":[596,814],"(1879\u20131929),":[597],"Im":[598],"Shahar,":[599],"Jerusalem:":[600],"Ha-Madpis,":[601],"1927":[602],"1,511":[603],"990019385920205171":[604],"Azriel":[605],"Natan":[606],"Frank":[607],"(1863\u20131924),":[608],"Sefer":[609],"Agadot":[610],"ha-Zohar,":[611],"Ahiasaf,":[613],"1923\u20131924":[614],"8":[615],"990013538860205171":[616],"Yitzhak":[617],"Katzenelson":[618],"(1885\u20131944),":[619],"Shirim,":[620],"Lodz:":[621],"Hovevei":[622],"Shira,":[623],"1938":[624],"1":[625],"Scans:":[626],"From":[627],"collections":[629],"Israel":[635],"/":[636,1072,1080],"\u05de\u05d0\u05d5\u05e1\u05e3":[637],"\u05d4\u05e1\u05e4\u05e8\u05d9\u05d9\u05d4":[638],"\u05d4\u05dc\u05d0\u05d5\u05de\u05d9\u05ea.":[639],"record's":[641],"IIIF":[642,656],"manifest":[643,1126],"carries":[644],"NLI's":[645,655],"licence.":[650],"fetched":[653],"service.":[657],"Google-digitized":[659],"\"non-commercial":[660],"scan":[662],"used.":[664],"Text:":[665],"edition":[674],"(public":[675],"domain":[676,837],"dump,":[677],"release":[678,1090],"2026-03),":[679],"placed":[680],"each":[682,965,1013,1443],"alignment":[687,697],"tool":[688],"(hocrgen":[689,1125,1169],"F8f).":[690],"kept":[693,983],"where":[695,1436],"scored":[698,1078,1131,1147],"least":[700,717],"0.9":[701],"agreed":[705,1441],"reading":[708,1381],"similarity":[714],"0.7.":[718],"located":[721],"chose":[725],"between":[726],"candidate":[727],"spans":[728],"line's":[731],"edges;":[732],"label":[734,966],"Label":[738],"normalisation:":[739],"Unicode":[740],"NFC,":[741],"removed;":[745],"\u2026":[746,995],"written":[747,968],"...,":[749],"\"":[750],"\",":[752],"soft":[753],"hyphens":[754],"removed.":[755,791],"contain":[759,769],"niqqud":[761,790],"letters;":[765],"90":[766],"them":[768,1149],"digits.":[770],"Pointed":[771],"prints":[772],"dropped:":[773],"two":[774,869],"fully":[775],"(Mahamadim,":[778],"Mandelkern)":[779],"left":[781,984,1377],"out.":[782],"Shirim":[783,852],"keeps":[784],"short":[786],"Edition":[792],"spelling:":[793],"Ben-Yehuda's":[794],"sometimes":[796],"differs":[797],"(for":[801],"\u05d0\u05ea\u05d4":[803],"\u05d0\u05d5\u05ea\u05d4).":[807],"Such":[808],"variants":[809],"labels;":[813],"has":[815,856],"most,":[817],"5.7%":[819],"words.":[821],"All":[822],"authors":[823],"died":[824],"more":[825],"70":[827,944],"years":[828],"ago,":[829],"so":[830,1411,1450],"public":[836],"Israel.":[839],"four":[841],"before":[844],"1931":[845],"also":[847],"past":[848],"US":[850,859],"term.":[851],"(1938,":[853],"line)":[855],"established":[858],"status.":[860],"Selection":[861],"(not":[863],"on):":[865],"7,092":[866],"further":[870],"held":[877,1157],"whole:":[879],"Kadish":[880],"Yehuda":[881],"Silman":[882],"(1880\u20131937),":[883],"Sansinim,":[884],"Tarbut,":[885],"1929":[886],"(990020250940205171,":[887],"3,046":[888],"lines),":[889],"Reuven":[891],"Brainin":[892],"(1862\u20131939),":[893],"Avraham":[894],"Mapu:":[895],"Hayav":[896],"u-Sfarav,":[897],"1928":[899],"(990018546440205171,":[900],"lines).":[902],"procedure":[904],"hyperparameters":[906],"Start:":[907],"(e12c65a),":[910],"extracted":[913],"combine_tessdata.":[915],"network":[917],"base's":[920,950],"unchanged:":[922],"[1,36,0,1Ct3,3,16Mp3,3Lfys48Lfx96Lrx96Lfx192O1c1].":[923],"Character":[924],"set:":[925],"base's,":[927],"plus":[928],"maqaf":[930],"\u05be":[931],"en":[934],"dash":[935],"\u2013.":[936],"output":[938],"layer":[939],"grew":[940],"69":[942],"codes.":[945],"dictionary":[947],"word":[952],"list,":[953],"rebuilt":[954],"build":[957],"langdata_lstm":[960],"07930fd":[961],"(Apache-2.0).":[962],"Truth":[963],"order:":[964,973],"reversed,":[977],"punctuation":[978],"mirrored,":[980],"digit":[981],"right.":[986],"converted":[990],".lstmf":[992],"lstm.train.":[998],"Trainer:":[999],"lstmtraining":[1000,1083],"--continue_from":[1001],"base,":[1003,1023,1357],"5.5.3,":[1005,1182],"CPU.":[1007],"Two":[1008],"arms":[1009],"100,000":[1011],"iterations":[1012],"(--target_error_rate":[1014],"0.01):":[1015],"lrdef:":[1016],"learning":[1018],"stored":[1020],"command":[1028],"line;":[1029],"lr1e4:":[1030],"--learning_rate":[1031],"0.0001":[1032],"--reset_learning_rate.":[1033],"Selection:":[1034],"checkpoint":[1036,1077,1141],"lowest":[1039],"BCER":[1040,1061],"(lstmeval)":[1041],"selection":[1044],"lines,":[1045],"over":[1046,1173],"all":[1047,1340],"96":[1048],"saved":[1049],"checkpoints":[1050],"both":[1052],"arms.":[1053],"That":[1054,1460],"lrdef":[1056],"iteration":[1058],"54,500":[1059],"(heb_0.603_16919_54500):":[1060],"2.03%,":[1062],"BWER":[1063],"5.30%.":[1064],"scores":[1070],"11.45%":[1071],"23.34%.":[1073],"best":[1075],"lr1e4":[1076],"2.14%":[1079],"5.81%.":[1081],"Conversion:":[1082],"--stop_training":[1084],"heb.traineddata,":[1086],"renamed":[1087],"(same":[1091],"bytes).":[1092],"One":[1093],"run.":[1095],"deterministic,":[1099],"reproducible.":[1105],"Evaluation":[1106],"Real":[1107],"Testing":[1109],"data:":[1110],"E2d4c,":[1111],"14":[1118],"Wikisource":[1120,1195,1316],"Indexes,":[1121],"e2d4c_wikisource_printed_v7).":[1127],"under":[1133,1165],"heocr-engine":[1134],"decision":[1135],"0012:":[1136],"training,":[1138],"tuning":[1139],"choice":[1142],"used":[1143],"them,":[1144],"once":[1150],"after":[1151],"selection,":[1152],"works":[1155],"training.":[1160],"Metric:":[1161],"WER":[1164,1285],"hocrgen's":[1166],"primary.v1":[1167],"normalisation":[1168],"d4a3414":[1170],"evaluate-references),":[1171],"pooled":[1172],"pages.":[1174,1308],"Every":[1175],"engine":[1176,1440],"reads":[1177],"whole":[1178],"3.":[1184],"Intervals:":[1185],"95%":[1186],"paired":[1187,1467],"bootstrap":[1188],"against":[1189,1290,1337],"Tesseract,":[1191],"resampling":[1192],"then":[1194],"Indexes":[1196,1317],"(clusters);":[1197],"2,000":[1198],"resamples,":[1199],"seed":[1200],"20261106.":[1201],"Engine":[1202],"CER,":[1203,1209],"WER,":[1206,1212],"Stock":[1215,1324],"(Homebrew)":[1218],"0.473":[1220],"0.384":[1221],"0.588":[1222],"(R6's":[1225],"base)":[1226],"0.225":[1227],"0.481":[1228],"0.376":[1229],"0.609":[1230],"0.209":[1232],"0.407":[1233],"0.383":[1234],"0.585":[1235],"minus":[1237],"\u0394CER,":[1238,1243],"(page;":[1241,1246],"cluster)":[1242,1247],"\u22120.016":[1250],"[\u22120.023,":[1251],"\u22120.010];":[1252],"cl":[1253,1259,1266,1272],"[\u22120.024,":[1254,1264],"\u22120.005]":[1255],"\u22120.002":[1256],"[\u22120.012,":[1257],"+0.012];":[1258],"[\u22120.006,":[1260,1270],"+0.002]":[1261],"\u22120.015":[1263],"\u22120.007];":[1265],"[\u22120.026,":[1267],"+0.001]":[1268],"+0.007":[1269],"+0.018];":[1271],"[\u22120.005,":[1273],"+0.019]":[1274],"0.016":[1283],"0.066":[1287],"[\u22120.086,":[1288],"\u22120.046]":[1289],"Tesseract.":[1292],"improves":[1294],"26":[1295],"leaves":[1297],"2":[1298],"unchanged":[1299],"worsens":[1301],"10.":[1302],"gain":[1304],"10-page":[1310],"subset":[1311],"whose":[1315],"fed":[1318],"glyph":[1322],"banks.":[1323],"highest":[1328],"there,":[1329],"level":[1333],"(0.383":[1336],"0.384).":[1338],"Almost":[1339],"R6's":[1342],"three":[1347],"likely":[1352],"reason.":[1353],"Against":[1354],"interval":[1360,1370],"excludes":[1361],"zero":[1362],"cluster":[1369],"just":[1371],"reaches":[1372],"(+0.001).":[1374],"What":[1375],"recognition,":[1379],"An":[1383],"order-free":[1384],"character-bag":[1385],"0.215":[1388],"(stock:":[1391,1408],"0.238),":[1392],"CER.":[1398],"emits":[1400],"7%":[1402],"fewer":[1403],"characters":[1404],"3%),":[1410],"deletions":[1412],"part":[1414],"error.":[1417],"Reference":[1418],"caveat:":[1419],"E2d4c":[1420],"human":[1423],"audit":[1424,1431],"earlier":[1427],"E2d3d":[1428],"references.":[1429],"Some":[1430],"proposals":[1432],"came":[1433],"places":[1435],"reference,":[1449],"corrections":[1452],"may":[1453],"lean":[1454],"slightly":[1455],"toward":[1456],"what":[1457],"reads.":[1459],"why":[1462],"every":[1463],"comparison":[1464],"above":[1465],"reference":[1471]},"counts_by_year":[],"updated_date":"2026-11-11T12:16:26.678958","created_date":"2026-11-11T00:00:00"}