{"id":369702,"date":"2026-08-17T17:07:18","date_gmt":"2026-08-17T11:37:18","guid":{"rendered":"https:\/\/forumias.com\/blog\/?p=369702"},"modified":"2026-08-17T17:07:18","modified_gmt":"2026-08-17T11:37:18","slug":"sravaani","status":"publish","type":"post","link":"https:\/\/forumias.com\/blog\/sravaani\/","title":{"rendered":"SraVaani"},"content":{"rendered":"<div class=\"content-box-green\">\n<p><strong>News:<\/strong> IISc researchers, with ARTPARK and Google, have released SraVaani, a multilingual speech recognition model for Indian languages.<\/p>\n<\/div>\n<h2 class=\"red-h2-box\">About SraVaani<\/h2>\n<figure id=\"attachment_369710\" aria-describedby=\"caption-attachment-369710\" style=\"width: 378px\" class=\"wp-caption aligncenter\"><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" class=\" wp-image-369710\" src=\"https:\/\/i0.wp.com\/forumias.com\/blog\/wp-content\/uploads\/2026\/08\/Source-IISc.jpg?resize=378%2C369&#038;ssl=1\" alt=\"SraVaani\" width=\"378\" height=\"369\" srcset=\"https:\/\/i0.wp.com\/forumias.com\/blog\/wp-content\/uploads\/2026\/08\/Source-IISc.jpg?resize=300%2C293&amp;ssl=1 300w, https:\/\/i0.wp.com\/forumias.com\/blog\/wp-content\/uploads\/2026\/08\/Source-IISc.jpg?resize=768%2C750&amp;ssl=1 768w, https:\/\/i0.wp.com\/forumias.com\/blog\/wp-content\/uploads\/2026\/08\/Source-IISc.jpg?w=849&amp;ssl=1 849w\" sizes=\"auto, (max-width: 378px) 100vw, 378px\" \/><figcaption id=\"caption-attachment-369710\" class=\"wp-caption-text\">Source &#8211; IISc<\/figcaption><\/figure>\n<ul>\n<li><strong>SraVaani<\/strong>\u00a0is a <strong>multilingual Indian speech recognition model<\/strong> designed to <strong>support regional and non-scheduled Indian languages<\/strong>.<\/li>\n<li><strong>Developed by:<\/strong> It was developed by <strong>researchers at the Indian Institute of Science (IISc)\u2019s SPIRE Lab<\/strong>, in collaboration <strong>with ARTPARK.<\/strong><\/li>\n<li><strong>Supported by:<\/strong> <strong>Google<\/strong> supported its development.<\/li>\n<li><strong>Based on model:<\/strong> SraVaani is <strong>based on Project Vaani,<\/strong> one of IISc\u2019s major programmes to understand the linguistic diversity of India.<\/li>\n<li><strong>Key features:<\/strong>\n<ul>\n<li><strong>Speech recognition architecture:<\/strong> It uses a <strong>FastConformer-based architecture<\/strong> for automatic speech recognition.<\/li>\n<li><strong>Wide language coverage:<\/strong> It covers <strong>20 scheduled languages<\/strong> and <strong>45 regional languages<\/strong> and dialects, including <strong>Garo, Angika, Chakma, Kokborok, Tulu, Bundeli and Bajjika.<\/strong><\/li>\n<li><strong>Pan-India coverage:<\/strong> It includes languages from the <strong>Northeast, eastern, western, northern, southern and central regions<\/strong>, along with <strong>English and Sanskrit.<\/strong><\/li>\n<li><strong>Strong performance:<\/strong> It was evaluated on <strong>eight public benchmark datasets<\/strong> and achieved the <strong>lowest average word error rate among the systems evaluated.<\/strong>\n<ul>\n<li>For Garo, it recorded a 9.5% word error rate, compared with 69.4% for the next-best system.<\/li>\n<\/ul>\n<\/li>\n<li><strong>Open access:<\/strong> It is <strong>freely available on Hugging Face<\/strong> under an <strong>MIT licence,<\/strong> along with a demo and fine-tuning code.<\/li>\n<li><strong>Wider applications:<\/strong> SraVaani <strong>can support further work<\/strong> in <strong>regional-language AI<\/strong>, low-resource <strong>speech-to-text,<\/strong> <strong>automatic language detection<\/strong>, dialect recognition and <strong>sovereign AI applications.<\/strong><\/li>\n<\/ul>\n<\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>News: IISc researchers, with ARTPARK and Google, have released SraVaani, a multilingual speech recognition model for Indian languages. About SraVaani SraVaani\u00a0is a multilingual Indian speech recognition model designed to support regional and non-scheduled Indian languages. Developed by: It was developed by researchers at the Indian Institute of Science (IISc)\u2019s SPIRE Lab, in collaboration with ARTPARK.&hellip; <a class=\"more-link\" href=\"https:\/\/forumias.com\/blog\/sravaani\/\">Continue reading <span class=\"screen-reader-text\">SraVaani<\/span><\/a><\/p>\n","protected":false},"author":10401,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"jetpack_post_was_ever_published":false,"footnotes":""},"categories":[1566,1760,956,12039],"tags":[11872],"class_list":["post-369702","post","type-post","status-publish","format-standard","hentry","category-daily-factly-articles","category-polity-and-nation","category-schemes-and-programs","category-knolls","tag-9pm-daily-factly","entry"],"jetpack_featured_media_url":"","views":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/forumias.com\/blog\/wp-json\/wp\/v2\/posts\/369702","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/forumias.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/forumias.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/forumias.com\/blog\/wp-json\/wp\/v2\/users\/10401"}],"replies":[{"embeddable":true,"href":"https:\/\/forumias.com\/blog\/wp-json\/wp\/v2\/comments?post=369702"}],"version-history":[{"count":0,"href":"https:\/\/forumias.com\/blog\/wp-json\/wp\/v2\/posts\/369702\/revisions"}],"wp:attachment":[{"href":"https:\/\/forumias.com\/blog\/wp-json\/wp\/v2\/media?parent=369702"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/forumias.com\/blog\/wp-json\/wp\/v2\/categories?post=369702"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/forumias.com\/blog\/wp-json\/wp\/v2\/tags?post=369702"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}