{"id":112482,"date":"2023-12-24T21:10:24","date_gmt":"2023-12-24T19:10:24","guid":{"rendered":"https:\/\/prohoster.info\/blog\/novosti-interneta\/obnovlenie-golosovyh-dannyh-mozilla-common-voice-16-0"},"modified":"2023-12-24T21:10:24","modified_gmt":"2023-12-24T19:10:24","slug":"obnovlenie-golosovyh-dannyh-mozilla-common-voice-16-0","status":"publish","type":"post","link":"https:\/\/prohoster.info\/et\/blog\/news\/obnovlenie-golosovyh-dannyh-mozilla-common-voice-16-0","title":{"rendered":"Mozilla Common Voice 16.0 h\u00e4\u00e4ldataandmete ajakohandus","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"
Mozilla ettev\u00f5te v\u00e4rskendas Common Voice h\u00e4\u00e4ldata komplekte, mis sisaldavad \u00fcle 200 000 inimese h\u00e4\u00e4ldusn\u00e4ite. Andmed on avaldatud avaliku omandina (CC0). Pakutavaid komplekte saab kasutada masin\u00f5ppe s\u00fcsteemides k\u00f5ne ja s\u00fcnteesi mudelite loomiseks. Eelmise v\u00e4rskendusega v\u00f5rreldes on k\u00f5nematerjali maht kollektsioonis suurenenud 28,7-st 30,3 tuhandeni, millest 19,7 tuhat tundi on l\u00e4binud kontrollimise protsessi. Toetatud keelte arv on suurenenud 114-lt 120-le (lisasid jidi\u0161i, l\u00e4\u00e4ne-l\u00e4tikeelse, liigurikeelse, osseedi, telugu ja l\u00e4\u00e4ne-sierra-pueblase nahuatli). <\/p>\n
In preparing materials in English, 90.67 thousand people participated, dictating 3438 hours of speech (there were 88.9 thousand participants and 3347 hours). The set for the Belarusian language includes 8249 participants and 1641 hours of speech material (previously there were 8205 participants and 1632 hours), for the Russian language - 3133 participants and 265 hours (previously there were 3053 participants and 260 hours), for the Uzbek language - 2151 participants and 264 hours (previously there were 2141 participants and 263 hours), and for the Ukrainian language - 1058 participants and 108 hours (previously there were 1024 participants and 105 hours). <\/p>\n
Common Voice projekt on suunatud koost\u00f6\u00f6 korraldamisele h\u00e4\u00e4lemustrite andmebaasi kogumiseks, mis arvestab k\u00f5igi h\u00e4\u00e4lt\u00fc\u00fcride ja k\u00f5neviisidega. Kasutajatele pakutakse ekraanile kuvatud fraaside h\u00e4\u00e4ldamist v\u00f5i teiste kasutajate lisatud andmete kvaliteedi hindamist. Kogutud andmebaasi, mis sisaldab erineva h\u00e4\u00e4ldusega t\u00fc\u00fcpfraase inimk\u00f5nest, saab piiranguteta kasutada masin\u00f5ppe s\u00fcsteemides ja teadusprojektides. \n Allikas: opennet.ru<\/a> <\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"