{"id":106128,"date":"2022-12-16T15:36:45","date_gmt":"2022-12-16T13:36:47","guid":{"rendered":"https:\/\/prohoster.info\/blog\/novosti-interneta\/sistema-mashinnogo-obucheniya-stable-diffusion-adaptirovana-dlya-sinteza-muzyki"},"modified":"2022-12-16T15:36:45","modified_gmt":"2022-12-16T13:36:47","slug":"sistema-mashinnogo-obucheniya-stable-diffusion-adaptirovana-dlya-sinteza-muzyki","status":"publish","type":"post","link":"https:\/\/prohoster.info\/cs\/blog\/news\/sistema-mashinnogo-obucheniya-stable-diffusion-adaptirovana-dlya-sinteza-muzyki","title":{"rendered":"Syst\u00e9m strojov\u00e9ho u\u010den\u00ed Stable Diffusion byl p\u0159izp\u016fsoben pro synt\u00e9zu hudby","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

Projekt Riffusion vyv\u00edj\u00ed variantu syst\u00e9mu strojov\u00e9ho u\u010den\u00ed Stable Diffusion, p\u0159izp\u016fsobenou pro generov\u00e1n\u00ed hudby nam\u00edsto obr\u00e1zk\u016f. Hudba m\u016f\u017ee b\u00fdt syntetizov\u00e1na z textov\u00e9ho popisu v p\u0159irozen\u00e9m jazyce nebo na z\u00e1klad\u011b navr\u017een\u00e9 \u0161ablony. Slo\u017eky pro synt\u00e9zu hudby jsou naps\u00e1ny v jazyce Python s vyu\u017eit\u00edm frameworku PyTorch a jsou dostupn\u00e9 pod licenc\u00ed MIT. Obal s rozhran\u00edm je realizov\u00e1n v jazyce TypeScript a tak\u00e9 se distribuuje pod licenc\u00ed MIT. Natr\u00e9novan\u00e9 modely jsou otev\u0159eny pod permisivn\u00ed licenc\u00ed Creative ML OpenRAIL-M, kter\u00e1 umo\u017e\u0148uje vyu\u017eit\u00ed pro komer\u010dn\u00ed \u00fa\u010dely. <\/p>\n

Projekt je zaj\u00edmav\u00fd t\u00edm, \u017ee nad\u00e1le pou\u017e\u00edv\u00e1 pro generov\u00e1n\u00ed hudby modely \u201ez textu na obraz\u201c a \u201ez obrazu na obraz\u201c, ale jako obr\u00e1zky manipuluje se spektrogramy. Jin\u00fdmi slovy, klasick\u00e1 Stable Diffusion nen\u00ed tr\u00e9nov\u00e1na na fotografi\u00edch a obr\u00e1zc\u00edch, ale na spektrogramov\u00fdch obraz\u00edch, kter\u00e9 odr\u00e1\u017eej\u00ed zm\u011bnu frekvence a amplitudy zvukov\u00e9 vlny v \u010dase. Odpov\u00eddaj\u00edc\u00edm zp\u016fsobem se na v\u00fdstupu tak\u00e9 vytv\u00e1\u0159\u00ed spektrogram, kter\u00fd se n\u00e1sledn\u011b p\u0159etv\u00e1\u0159\u00ed na zvukovou reprezentaci.

\"Syst\u00e9m<\/center> <\/p>\n

Metoda m\u016f\u017ee b\u00fdt tak\u00e9 pou\u017eita ke zm\u011bn\u011b existuj\u00edc\u00edch zvukov\u00fdch skladeb a synt\u00e9ze hudby podle vzoru, obdobn\u011b jako se prov\u00e1d\u00ed modifikace obr\u00e1zk\u016f v Stable Diffusion. Nap\u0159\u00edklad p\u0159i generaci mohou b\u00fdt zad\u00e1ny vzory spektrogram\u016f s referen\u010dn\u00edm stylem, kombinov\u00e1ny r\u016fzn\u00e9 styly, prov\u00e1d\u011bny plynul\u00e9 p\u0159echody z jednoho stylu na druh\u00fd nebo prov\u00e1d\u011bny \u00fapravy existuj\u00edc\u00edho zvuku k \u0159e\u0161en\u00ed takov\u00fdch \u00fakol\u016f, jako je zv\u00fd\u0161en\u00ed hlasitosti jednotliv\u00fdch n\u00e1stroj\u016f, zm\u011bna rytmu a v\u00fdm\u011bna n\u00e1stroj\u016f. Vzory se tak\u00e9 pou\u017e\u00edvaj\u00ed pro generaci dlouhotrvaj\u00edc\u00edch skladeb, kter\u00e9 jsou kompozice z \u0159ady bl\u00edzk\u00fdch \u00faryvk\u016f, m\u00edrn\u011b se m\u011bn\u00edc\u00edch v \u010dase. Samostatn\u011b generovan\u00e9 \u00faryvky jsou spojov\u00e1ny do kontinu\u00e1ln\u00edho toku pomoc\u00ed interpolace vnit\u0159n\u00edch parametr\u016f modelu.

\"Syst\u00e9m<\/center> <\/p>\n

K vytvo\u0159en\u00ed spektrogramu ze zvuku se pou\u017e\u00edv\u00e1 ok\u00e9nkov\u00e1 Fourierova transformace. P\u0159i rekonstrukci zvuku ze spektrogramu vznik\u00e1 probl\u00e9m s ur\u010den\u00edm f\u00e1ze (ve spektrogramu jsou pouze frekvence a amplitudy), pro rekonstrukci kter\u00e9 je vyu\u017eit algoritmus aproximation Grifina-Lima.

<\/center>
<\/center>
\n
Zdroj: opennet.ru<\/a> <\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"

\u041f\u0440\u043e\u0435\u043a\u0442\u043e\u043c Riffusion \u0440\u0430\u0437\u0432\u0438\u0432\u0430\u0435\u0442\u0441\u044f \u0432\u0430\u0440\u0438\u0430\u043d\u0442 \u0441\u0438\u0441\u0442\u0435\u043c\u044b \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f Stable Diffusion, \u0430\u0434\u0430\u043f\u0442\u0438\u0440\u043e\u0432\u0430\u043d\u043d\u044b\u0439 \u0434\u043b\u044f \u0433\u0435\u043d\u0435\u0440\u0430\u0446\u0438\u0438 \u043c\u0443\u0437\u044b\u043a\u0438 \u0432\u043c\u0435\u0441\u0442\u043e \u0438\u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u0438\u0439. \u041c\u0443\u0437\u044b\u043a\u0430 \u043c\u043e\u0436\u0435\u0442 \u0441\u0438\u043d\u0442\u0435\u0437\u0438\u0440\u043e\u0432\u0430\u0442\u044c\u0441\u044f \u0442\u0435\u043a\u0441\u0442\u043e\u0432\u043e\u0433\u043e \u043e\u043f\u0438\u0441\u0430\u043d\u0438\u044f \u043d\u0430 \u0435\u0441\u0442\u0435\u0441\u0442\u0432\u0435\u043d\u043d\u043e\u043c \u044f\u0437\u044b\u043a\u0435 \u0438\u043b\u0438 \u043d\u0430 \u043e\u0441\u043d\u043e\u0432\u0435 \u043f\u0440\u0435\u0434\u043b\u043e\u0436\u0435\u043d\u043d\u043e\u0433\u043e \u0448\u0430\u0431\u043b\u043e\u043d\u0430. \u041a\u043e\u043c\u043f\u043e\u043d\u0435\u043d\u0442\u044b \u0434\u043b\u044f \u0441\u0438\u043d\u0442\u0435\u0437\u0430 \u043c\u0443\u0437\u044b\u043a\u0438 \u043d\u0430\u043f\u0438\u0441\u0430\u043d\u044b \u043d\u0430 \u044f\u0437\u044b\u043a\u0435 Python \u0441 \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u043d\u0438\u0435\u043c \u0444\u0440\u0435\u0439\u043c\u0432\u043e\u0440\u043a\u0430 PyTorch \u0438 \u0434\u043e\u0441\u0442\u0443\u043f\u043d\u044b \u043f\u043e\u0434 \u043b\u0438\u0446\u0435\u043d\u0437\u0438\u0435\u0439 MIT. \u041e\u0431\u0432\u044f\u0437\u043a\u0430 \u0441 \u0438\u043d\u0442\u0435\u0440\u0444\u0435\u0439\u0441\u043e\u043c \u0440\u0435\u0430\u043b\u0438\u0437\u043e\u0432\u0430\u043d\u0430 \u043d\u0430 \u044f\u0437\u044b\u043a\u0435 TypeScript \u0438 \u0442\u0430\u043a\u0436\u0435 \u0440\u0430\u0441\u043f\u0440\u043e\u0441\u0442\u0440\u0430\u043d\u044f\u0435\u0442\u0441\u044f […]<\/p>\n","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"author":1,"featured_media":106129,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[702],"tags":[],"class_list":["post-106128","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-news"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\n","aioseo_head_json":{"title":"\ud83e\udd47 Syst\u00e9m strojov\u00e9ho u\u010den\u00ed Stable Diffusion je p\u0159izp\u016fsoben pro synt\u00e9zu hudby | ProHoster","description":"Projekt Riffusion vyv\u00edj\u00ed variantu syst\u00e9mu strojov\u00e9ho u\u010den\u00ed Stable Diffusion, kter\u00fd je p\u0159izp\u016fsoben k generov\u00e1n\u00ed hudby m\u00edsto obr\u00e1zk\u016f.","canonical_url":"https:\/\/prohoster.info\/cs\/blog\/news\/sistema-mashinnogo-obucheniya-stable-diffusion-adaptirovana-dlya-sinteza-muzyki","robots":"max-image-preview:large","keywords":"","webmasterTools":{"miscellaneous":""},"schema":null,"og:locale":"cs_CZ","og:site_name":"ProHoster | \u041a\u0443\u043f\u0438\u0442\u044c \u043d\u0430\u0434\u0435\u0436\u043d\u044b\u0439 \u0445\u043e\u0441\u0442\u0438\u043d\u0433 \u0434\u043b\u044f \u0441\u0430\u0439\u0442\u043e\u0432 \u0441 \u0437\u0430\u0449\u0438\u0442\u043e\u0439 \u043e\u0442 DDoS, VPS VDS \u0441\u0435\u0440\u0432\u0435\u0440\u044b","og:type":"article","og:title":"\ud83e\udd47\u0421\u0438\u0441\u0442\u0435\u043c\u0430 \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f Stable Diffusion \u0430\u0434\u0430\u043f\u0442\u0438\u0440\u043e\u0432\u0430\u043d\u0430 \u0434\u043b\u044f \u0441\u0438\u043d\u0442\u0435\u0437\u0430 \u043c\u0443\u0437\u044b\u043a\u0438 | ProHoster","og:description":"\u041f\u0440\u043e\u0435\u043a\u0442\u043e\u043c Riffusion \u0440\u0430\u0437\u0432\u0438\u0432\u0430\u0435\u0442\u0441\u044f \u0432\u0430\u0440\u0438\u0430\u043d\u0442 \u0441\u0438\u0441\u0442\u0435\u043c\u044b \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f Stable Diffusion, \u0430\u0434\u0430\u043f\u0442\u0438\u0440\u043e\u0432\u0430\u043d\u043d\u044b\u0439 \u0434\u043b\u044f \u0433\u0435\u043d\u0435\u0440\u0430\u0446\u0438\u0438 \u043c\u0443\u0437\u044b\u043a\u0438 \u0432\u043c\u0435\u0441\u0442\u043e \u0438\u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u0438\u0439.","og:url":"https:\/\/prohoster.info\/cs\/blog\/news\/sistema-mashinnogo-obucheniya-stable-diffusion-adaptirovana-dlya-sinteza-muzyki","og:image":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:secure_url":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:width":350,"og:image:height":350,"article:published_time":"2022-12-16T13:36:47+00:00","article:modified_time":"2022-12-16T13:36:47+00:00","article:publisher":"https:\/\/www.facebook.com\/prohoster","article:author":"https:\/\/www.facebook.com\/prohoster"},"aioseo_meta_data":[],"gt_translate_keys":[{"key":"link","format":"url"}],"_links":{"self":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/posts\/106128","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/comments?post=106128"}],"version-history":[{"count":0,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/posts\/106128\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/media\/106129"}],"wp:attachment":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/media?parent=106128"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/categories?post=106128"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/tags?post=106128"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}