{"id":2636,"date":"2026-01-18T16:28:43","date_gmt":"2026-01-18T15:28:43","guid":{"rendered":"https:\/\/labalec.fr\/erwan\/?p=2636"},"modified":"2026-01-18T16:35:06","modified_gmt":"2026-01-18T15:35:06","slug":"whisper-built-for-vulkan","status":"publish","type":"post","link":"https:\/\/labalec.fr\/erwan\/?p=2636","title":{"rendered":"Whisper built for Vulkan"},"content":{"rendered":"\n<p>Find it <a href=\"https:\/\/erwan.labalec.fr\/1.8.3-vulkan.zip\">here<\/a>.<\/p>\n\n\n\n<p>This version will use your gpu and is about 3 to 4 times faster.<\/p>\n\n\n\n<p>Although quality is slightly lower compared to CPU.<\/p>\n\n\n\n<p>I am getting good results (i.e less hallucinations &amp; loops) thus with the below parameters.<\/p>\n\n\n\n<p>If the idea is to use the transcript to have an IA turn into minutes\/summary, it is plenty acceptable.<\/p>\n\n\n\n<p>In your prompt then, mention the transcript has been generated by an IA with possible hallucinations, loops, etc and also provide a bit of context (it was a meeting about X and Z with person\/function A and person\/function B) : it usually provides very good results.<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>whisper-cli.exe -m \"..\/ggml-base.bin\" -f \"..\/output.wav\" -osrt --language fr --max-context 50 --beam-size 3 --temperature 0 --temperature-inc 0.2 --threads 8 --split-on-word<\/code><\/pre>\n\n\n\n<p><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Find it here. This version will use your gpu and is about 3 to 4 times faster. Although quality is slightly lower compared to CPU. I am getting good results (i.e less hallucinations &amp; loops) thus with the below parameters. If the idea is to use the transcript to have an IA turn into minutes\/summary, <a href='https:\/\/labalec.fr\/erwan\/?p=2636' class='excerpt-more'>[&#8230;]<\/a><\/p>\n","protected":false},"author":7,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[150],"tags":[],"class_list":["post-2636","post","type-post","status-publish","format-standard","hentry","category-whisper","category-150-id","post-seq-1","post-parity-odd","meta-position-corners","fix"],"_links":{"self":[{"href":"https:\/\/labalec.fr\/erwan\/index.php?rest_route=\/wp\/v2\/posts\/2636","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/labalec.fr\/erwan\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/labalec.fr\/erwan\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/labalec.fr\/erwan\/index.php?rest_route=\/wp\/v2\/users\/7"}],"replies":[{"embeddable":true,"href":"https:\/\/labalec.fr\/erwan\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=2636"}],"version-history":[{"count":2,"href":"https:\/\/labalec.fr\/erwan\/index.php?rest_route=\/wp\/v2\/posts\/2636\/revisions"}],"predecessor-version":[{"id":2639,"href":"https:\/\/labalec.fr\/erwan\/index.php?rest_route=\/wp\/v2\/posts\/2636\/revisions\/2639"}],"wp:attachment":[{"href":"https:\/\/labalec.fr\/erwan\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=2636"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/labalec.fr\/erwan\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=2636"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/labalec.fr\/erwan\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=2636"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}