{"id":12194,"date":"2024-02-16T07:49:12","date_gmt":"2024-02-16T10:49:12","guid":{"rendered":"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/"},"modified":"2024-02-16T07:49:12","modified_gmt":"2024-02-16T10:49:12","slug":"amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes","status":"publish","type":"post","link":"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/","title":{"rendered":"Amazon treina LLM de 980 milh\u00f5es de par\u00e2metros com \u2018habilidades emergentes\u2019"},"content":{"rendered":"
\n

Researchers at Amazon have trained a new large language model (LLM) for text-to-speech that they claim exhibits \u201cemergent\u201d abilities.\u00a0<\/p>\n

The 980 million parameter model, called BASE TTS, is the largest text-to-speech model yet created. The researchers trained models of various sizes on up to 100,000 hours of public domain speech data to see if they would observe the same performance leaps that occur in natural language processing models once they grow past a certain scale.\u00a0<\/p>\n

They found that their medium-sized 400 million parameter model \u2013 trained on 10,000 hours of audio \u2013 showed a marked improvement in versatility and robustness on tricky test sentences.<\/p>\n

The test sentences contained complex lexical, syntactic, and paralinguistic features like compound nouns, emotions, foreign words, and punctuation that normally trip up text-to-speech systems. While BASE TTS did not handle them perfectly, it made significantly fewer errors in stress, intonation, and pronunciation than existing models.<\/p>\n

\u201cThese sentences are designed to contain challenging tasks\u2014none of which BASE TTS is explicitly trained to perform,\u201d explained the researchers.\u00a0<\/p>\n

The largest 980 million parameter version of the model \u2013 trained on 100,000 hours of audio \u2013 did not demonstrate further abilities beyond the 400 million parameter version.<\/p>\n

While an experimental process<\/p>\n","protected":false},"excerpt":{"rendered":"

Researchers at Amazon have trained a new large language model (LLM) for text-to-speech that they claim exhibits \u201cemergent\u201d abilities.\u00a0 The 980 million parameter model, called BASE TTS, is the largest text-to-speech model yet created. The researchers trained models of various sizes on up to 100,000 hours of public domain speech data to see if they…<\/p>\n","protected":false},"author":1,"featured_media":12195,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[30],"tags":[67],"class_list":["post-12194","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-inteligencia-artificial","tag-ia"],"yoast_head":"\nAmazon treina LLM de 980 milh\u00f5es de par\u00e2metros com \u2018habilidades emergentes\u2019 - 4tuneagencia.digital<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/\" \/>\n<meta property=\"og:locale\" content=\"pt_BR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Amazon treina LLM de 980 milh\u00f5es de par\u00e2metros com \u2018habilidades emergentes\u2019\" \/>\n<meta property=\"og:description\" content=\"Researchers at Amazon have trained a new large language model (LLM) for text-to-speech that they claim exhibits \u201cemergent\u201d abilities.\u00a0 The 980 million parameter model, called BASE TTS, is the largest text-to-speech model yet created. The researchers trained models of various sizes on up to 100,000 hours of public domain speech data to see if they...\" \/>\n<meta property=\"og:url\" content=\"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/\" \/>\n<meta property=\"og:site_name\" content=\"4tuneagencia.digital\" \/>\n<meta property=\"article:published_time\" content=\"2024-02-16T10:49:12+00:00\" \/>\n<meta name=\"author\" content=\"4tune\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Escrito por\" \/>\n\t<meta name=\"twitter:data1\" content=\"4tune\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. tempo de leitura\" \/>\n\t<meta name=\"twitter:data2\" content=\"1 minuto\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/\"},\"author\":{\"name\":\"4tune\",\"@id\":\"https:\/\/4tuneagencia.digital\/blog\/#\/schema\/person\/7c40d2e8b9964c2390122613459c2100\"},\"headline\":\"Amazon treina LLM de 980 milh\u00f5es de par\u00e2metros com \u2018habilidades emergentes\u2019\",\"datePublished\":\"2024-02-16T10:49:12+00:00\",\"dateModified\":\"2024-02-16T10:49:12+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/\"},\"wordCount\":203,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\/\/4tuneagencia.digital\/blog\/#organization\"},\"image\":{\"@id\":\"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/4tuneagencia.digital\/blog\/wp-content\/uploads\/2024\/02\/wp-header-logo-185.png\",\"keywords\":[\"IA\"],\"articleSection\":[\"Intelig\u00eancia artificial\"],\"inLanguage\":\"pt-BR\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/\",\"url\":\"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/\",\"name\":\"Amazon treina LLM de 980 milh\u00f5es de par\u00e2metros com \u2018habilidades emergentes\u2019 - 4tuneagencia.digital\",\"isPartOf\":{\"@id\":\"https:\/\/4tuneagencia.digital\/blog\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/4tuneagencia.digital\/blog\/wp-content\/uploads\/2024\/02\/wp-header-logo-185.png\",\"datePublished\":\"2024-02-16T10:49:12+00:00\",\"dateModified\":\"2024-02-16T10:49:12+00:00\",\"breadcrumb\":{\"@id\":\"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/#breadcrumb\"},\"inLanguage\":\"pt-BR\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"pt-BR\",\"@id\":\"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/#primaryimage\",\"url\":\"https:\/\/4tuneagencia.digital\/blog\/wp-content\/uploads\/2024\/02\/wp-header-logo-185.png\",\"contentUrl\":\"https:\/\/4tuneagencia.digital\/blog\/wp-content\/uploads\/2024\/02\/wp-header-logo-185.png\",\"width\":728,\"height\":90},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"In\u00edcio\",\"item\":\"https:\/\/4tuneagencia.digital\/blog\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Amazon treina LLM de 980 milh\u00f5es de par\u00e2metros com \u2018habilidades emergentes\u2019\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/4tuneagencia.digital\/blog\/#website\",\"url\":\"https:\/\/4tuneagencia.digital\/blog\/\",\"name\":\"4tuneagencia.digital\",\"description\":\"Marketing e Web Designer \",\"publisher\":{\"@id\":\"https:\/\/4tuneagencia.digital\/blog\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/4tuneagencia.digital\/blog\/?s={search_term_string}\"},\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"pt-BR\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/4tuneagencia.digital\/blog\/#organization\",\"name\":\"4tuneagencia.digital\",\"url\":\"https:\/\/4tuneagencia.digital\/blog\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"pt-BR\",\"@id\":\"https:\/\/4tuneagencia.digital\/blog\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/4tuneagencia.digital\/blog\/wp-content\/uploads\/2023\/09\/novo-logo-e1694954525541.png\",\"contentUrl\":\"https:\/\/4tuneagencia.digital\/blog\/wp-content\/uploads\/2023\/09\/novo-logo-e1694954525541.png\",\"width\":250,\"height\":250,\"caption\":\"4tuneagencia.digital\"},\"image\":{\"@id\":\"https:\/\/4tuneagencia.digital\/blog\/#\/schema\/logo\/image\/\"}},{\"@type\":\"Person\",\"@id\":\"https:\/\/4tuneagencia.digital\/blog\/#\/schema\/person\/7c40d2e8b9964c2390122613459c2100\",\"name\":\"4tune\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"pt-BR\",\"@id\":\"https:\/\/4tuneagencia.digital\/blog\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/e81e8ea0176022ee4555117aee5e3092bff2a7aaa930134f8023c7d4bc2f1495?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/e81e8ea0176022ee4555117aee5e3092bff2a7aaa930134f8023c7d4bc2f1495?s=96&d=mm&r=g\",\"caption\":\"4tune\"},\"sameAs\":[\"http:\/\/4tuneagencia.digital\"],\"url\":\"https:\/\/4tuneagencia.digital\/blog\/author\/admin\/\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Amazon treina LLM de 980 milh\u00f5es de par\u00e2metros com \u2018habilidades emergentes\u2019 - 4tuneagencia.digital","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/","og_locale":"pt_BR","og_type":"article","og_title":"Amazon treina LLM de 980 milh\u00f5es de par\u00e2metros com \u2018habilidades emergentes\u2019","og_description":"Researchers at Amazon have trained a new large language model (LLM) for text-to-speech that they claim exhibits \u201cemergent\u201d abilities.\u00a0 The 980 million parameter model, called BASE TTS, is the largest text-to-speech model yet created. The researchers trained models of various sizes on up to 100,000 hours of public domain speech data to see if they...","og_url":"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/","og_site_name":"4tuneagencia.digital","article_published_time":"2024-02-16T10:49:12+00:00","author":"4tune","twitter_card":"summary_large_image","twitter_misc":{"Escrito por":"4tune","Est. tempo de leitura":"1 minuto"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/#article","isPartOf":{"@id":"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/"},"author":{"name":"4tune","@id":"https:\/\/4tuneagencia.digital\/blog\/#\/schema\/person\/7c40d2e8b9964c2390122613459c2100"},"headline":"Amazon treina LLM de 980 milh\u00f5es de par\u00e2metros com \u2018habilidades emergentes\u2019","datePublished":"2024-02-16T10:49:12+00:00","dateModified":"2024-02-16T10:49:12+00:00","mainEntityOfPage":{"@id":"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/"},"wordCount":203,"commentCount":0,"publisher":{"@id":"https:\/\/4tuneagencia.digital\/blog\/#organization"},"image":{"@id":"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/#primaryimage"},"thumbnailUrl":"https:\/\/4tuneagencia.digital\/blog\/wp-content\/uploads\/2024\/02\/wp-header-logo-185.png","keywords":["IA"],"articleSection":["Intelig\u00eancia artificial"],"inLanguage":"pt-BR","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/#respond"]}]},{"@type":"WebPage","@id":"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/","url":"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/","name":"Amazon treina LLM de 980 milh\u00f5es de par\u00e2metros com \u2018habilidades emergentes\u2019 - 4tuneagencia.digital","isPartOf":{"@id":"https:\/\/4tuneagencia.digital\/blog\/#website"},"primaryImageOfPage":{"@id":"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/#primaryimage"},"image":{"@id":"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/#primaryimage"},"thumbnailUrl":"https:\/\/4tuneagencia.digital\/blog\/wp-content\/uploads\/2024\/02\/wp-header-logo-185.png","datePublished":"2024-02-16T10:49:12+00:00","dateModified":"2024-02-16T10:49:12+00:00","breadcrumb":{"@id":"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/#breadcrumb"},"inLanguage":"pt-BR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/"]}]},{"@type":"ImageObject","inLanguage":"pt-BR","@id":"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/#primaryimage","url":"https:\/\/4tuneagencia.digital\/blog\/wp-content\/uploads\/2024\/02\/wp-header-logo-185.png","contentUrl":"https:\/\/4tuneagencia.digital\/blog\/wp-content\/uploads\/2024\/02\/wp-header-logo-185.png","width":728,"height":90},{"@type":"BreadcrumbList","@id":"https:\/\/4tuneagencia.digital\/blog\/amazon-treina-llm-de-980-milhoes-de-parametros-com-habilidades-emergentes\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"In\u00edcio","item":"https:\/\/4tuneagencia.digital\/blog\/"},{"@type":"ListItem","position":2,"name":"Amazon treina LLM de 980 milh\u00f5es de par\u00e2metros com \u2018habilidades emergentes\u2019"}]},{"@type":"WebSite","@id":"https:\/\/4tuneagencia.digital\/blog\/#website","url":"https:\/\/4tuneagencia.digital\/blog\/","name":"4tuneagencia.digital","description":"Marketing e Web Designer ","publisher":{"@id":"https:\/\/4tuneagencia.digital\/blog\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/4tuneagencia.digital\/blog\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"pt-BR"},{"@type":"Organization","@id":"https:\/\/4tuneagencia.digital\/blog\/#organization","name":"4tuneagencia.digital","url":"https:\/\/4tuneagencia.digital\/blog\/","logo":{"@type":"ImageObject","inLanguage":"pt-BR","@id":"https:\/\/4tuneagencia.digital\/blog\/#\/schema\/logo\/image\/","url":"https:\/\/4tuneagencia.digital\/blog\/wp-content\/uploads\/2023\/09\/novo-logo-e1694954525541.png","contentUrl":"https:\/\/4tuneagencia.digital\/blog\/wp-content\/uploads\/2023\/09\/novo-logo-e1694954525541.png","width":250,"height":250,"caption":"4tuneagencia.digital"},"image":{"@id":"https:\/\/4tuneagencia.digital\/blog\/#\/schema\/logo\/image\/"}},{"@type":"Person","@id":"https:\/\/4tuneagencia.digital\/blog\/#\/schema\/person\/7c40d2e8b9964c2390122613459c2100","name":"4tune","image":{"@type":"ImageObject","inLanguage":"pt-BR","@id":"https:\/\/4tuneagencia.digital\/blog\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/e81e8ea0176022ee4555117aee5e3092bff2a7aaa930134f8023c7d4bc2f1495?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/e81e8ea0176022ee4555117aee5e3092bff2a7aaa930134f8023c7d4bc2f1495?s=96&d=mm&r=g","caption":"4tune"},"sameAs":["http:\/\/4tuneagencia.digital"],"url":"https:\/\/4tuneagencia.digital\/blog\/author\/admin\/"}]}},"jetpack_featured_media_url":"https:\/\/4tuneagencia.digital\/blog\/wp-content\/uploads\/2024\/02\/wp-header-logo-185.png","_links":{"self":[{"href":"https:\/\/4tuneagencia.digital\/blog\/wp-json\/wp\/v2\/posts\/12194","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/4tuneagencia.digital\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/4tuneagencia.digital\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/4tuneagencia.digital\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/4tuneagencia.digital\/blog\/wp-json\/wp\/v2\/comments?post=12194"}],"version-history":[{"count":0,"href":"https:\/\/4tuneagencia.digital\/blog\/wp-json\/wp\/v2\/posts\/12194\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/4tuneagencia.digital\/blog\/wp-json\/wp\/v2\/media\/12195"}],"wp:attachment":[{"href":"https:\/\/4tuneagencia.digital\/blog\/wp-json\/wp\/v2\/media?parent=12194"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/4tuneagencia.digital\/blog\/wp-json\/wp\/v2\/categories?post=12194"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/4tuneagencia.digital\/blog\/wp-json\/wp\/v2\/tags?post=12194"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}