Kompania Google prezantoi nj\u00eb kodek audio t\u00eb ri Lyra, i optimizuar p\u00ebr t\u00eb arritur cil\u00ebsin\u00eb maksimale t\u00eb transmetimit t\u00eb z\u00ebrit edhe n\u00eb kanale shum\u00eb t\u00eb ngadalta t\u00eb komunikimit. Kodi i implementimit t\u00eb Lyra \u00ebsht\u00eb shkruar n\u00eb C++ dhe \u00ebsht\u00eb i hapur n\u00ebn licenc\u00ebn Apache 2.0, por n\u00eb mesin e var\u00ebsive t\u00eb nevojshme p\u00ebr funksionimin e tij ndodhet nj\u00eb bibliotek\u00eb pron\u00ebsore libsparse_inference.so q\u00eb ka implementimin e b\u00ebrtham\u00ebs p\u00ebr llogaritjet matematikore. Vlen t\u00eb theksohet se biblioteca pron\u00ebsore \u00ebsht\u00eb p\u00ebrkohshme \u2014 m\u00eb von\u00eb, Google premton t\u00eb zhvilloj\u00eb nj\u00eb z\u00ebvend\u00ebsim t\u00eb hapur dhe t\u00eb siguroj\u00eb mb\u00ebshtetje p\u00ebr platforma t\u00eb ndryshme. <\/p>\n
N\u00eb aspektin e cil\u00ebsis\u00eb s\u00eb t\u00eb dh\u00ebnave t\u00eb z\u00ebrit t\u00eb transmetuara n\u00eb shpejt\u00ebsi t\u00eb ul\u00ebta, Lyra e tejkalon ndjesh\u00ebm kodek\u00ebt tradicional\u00eb q\u00eb p\u00ebrdorin metoda t\u00eb p\u00ebrpunimit digjital t\u00eb sinjaleve. P\u00ebr t\u00eb arritur nj\u00eb cil\u00ebsi t\u00eb lart\u00eb t\u00eb transmetimit t\u00eb z\u00ebrit n\u00eb kushte p\u00ebr nj\u00eb volum t\u00eb kufizuar t\u00eb informacionit t\u00eb transmetuar, p\u00ebrve\u00e7 metodave t\u00eb zakonshme t\u00eb kompresimit t\u00eb z\u00ebrit dhe transformimit t\u00eb sinjaleve, n\u00eb Lyra p\u00ebrdoret nj\u00eb model fjalimi i bazuar n\u00eb nj\u00eb sistem t\u00eb m\u00ebsimit t\u00eb makineris\u00eb, q\u00eb lejon rikrijimin e informacionit t\u00eb munguar mbi baz\u00ebn e karakteristikave tipike t\u00eb fjal\u00ebs. Modeli i angazhuar p\u00ebr gjenerimin e z\u00ebrit \u00ebsht\u00eb trajnuar duke p\u00ebrdorur disa mij\u00ebra or\u00eb regjistrimesh z\u00ebri n\u00eb m\u00eb shum\u00eb se 70 gjuh\u00eb.
<\/center> <\/p>\n
Kodi p\u00ebrfshin nj\u00eb kodues dhe nj\u00eb dekodues. Algoritmi i pun\u00ebs s\u00eb koduesit konsiston n\u00eb nxjerrjen e parametrave t\u00eb t\u00eb dh\u00ebnave audio \u00e7do 40 milisekonda, kompresimin e tyre dhe d\u00ebrgimin e tyre tek marr\u00ebsi p\u00ebrmes rrjetit. P\u00ebr t\u00eb transmetuar t\u00eb dh\u00ebnat, mjafton nj\u00eb kanal komunikimi me shpejt\u00ebsi 3 kilobit n\u00eb sekond\u00eb. Parametrat audio t\u00eb nxjerr\u00eb p\u00ebrfshijn\u00eb spektrogramet mel-logaritmike, q\u00eb marrin parasysh karakteristikat e energjis\u00eb s\u00eb z\u00ebrit n\u00eb gamat e ndryshme t\u00eb frekuenc\u00ebs dhe jan\u00eb p\u00ebrgatitur duke marr\u00eb parasysh modelin e perceptimit t\u00eb d\u00ebgjimit t\u00eb njeriut.
<\/center> <\/p>\n
Dekoduesi p\u00ebrdor nj\u00eb model gjenerativ, i cili, n\u00eb baz\u00eb t\u00eb parametrave audio t\u00eb d\u00ebrguar, rikrijon sinjalin me z\u00eb. P\u00ebr t\u00eb reduktuar kompleksitetin e llogaritjeve, \u00ebsht\u00eb aplikuar nj\u00eb model i leht\u00eb i bazuar n\u00eb nj\u00eb rrjet nervor rekurent, q\u00eb p\u00ebrfaq\u00ebson nj\u00eb variant t\u00eb modelit t\u00eb sintez\u00ebs s\u00eb z\u00ebrit WaveRNN, n\u00eb t\u00eb cilin p\u00ebrdoret nj\u00eb frekuenc\u00eb m\u00eb e ul\u00ebt e mostrave, por gjithashtu gjenerohen paralelisht disa sinjale n\u00eb gam\u00eb t\u00eb ndryshme frekuencash. Sinjalet e marra pastaj mbivendosen p\u00ebr t\u00eb marr\u00eb nj\u00eb sinjal t\u00eb vet\u00ebm t\u00eb daljes, q\u00eb i p\u00ebrgjigjet frekuenc\u00ebs s\u00eb caktuar t\u00eb diskretizimit. <\/p>\n
P\u00ebrshpejtimi gjithashtu p\u00ebrfshin instrukcione t\u00eb specializuara t\u00eb procesor\u00ebve, t\u00eb disponueshme n\u00eb procesor\u00ebt 64-bit ARM. Si rezultat, pavar\u00ebsisht p\u00ebrdorimit t\u00eb m\u00ebsimit makinerik, kodeku Lyra mund t\u00eb p\u00ebrdoret p\u00ebr kodimin dhe dekodimin e z\u00ebrit n\u00eb koh\u00eb reale n\u00eb smartphone-t e nivelit t\u00eb mes\u00ebm, duke treguar nj\u00eb vones\u00eb t\u00eb transmetimit t\u00eb sinjalit prej 90 milisekondash. \n Burimi: opennet.ru<\/a> <\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"
![\"Google](\"\/wp-content\/uploads\/2021\/04\/dff1cbc3a2422197df2b4510b0a4fc75.png\")
<\/center> <\/p>\n![\"Google](\"\/wp-content\/uploads\/2021\/04\/5fedbeb0d9e364c54c9fd35dab76f019.png\")
<\/center> <\/p>\n