{"id":339289,"date":"2025-06-26T07:11:20","date_gmt":"2025-06-26T14:11:20","guid":{"rendered":"https:\/\/cms-articles.softonic.io\/en\/?p=339289"},"modified":"2025-07-01T14:17:41","modified_gmt":"2025-07-01T21:17:41","slug":"openai-charges-for-every-minute-its-ai-transcribes-so-they-have-found-a-way-to-fool-the-system","status":"publish","type":"post","link":"https:\/\/cms-articles.softonic.io\/en\/openai-charges-for-every-minute-its-ai-transcribes-so-they-have-found-a-way-to-fool-the-system\/","title":{"rendered":"OpenAI charges for every minute its AI transcribes, so they have found a way to fool the system"},"content":{"rendered":"\n<p class=\"wp-block-paragraph\">OpenAI\u2019s pay-as-you-go transcription looks straightforward: longer recordings mean more tokens and higher bills.&nbsp;<strong>Developer George Mandis stumbled on a way to shrink both time and cost dramatically<\/strong>&nbsp;by literally making the minutes shorter.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">A simple speed hack<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Mandis fed his audio through FFmpeg\u2019s atempo filter, tripling playback speed without pitch distortion.&nbsp;<strong>The accelerated track remained clear enough<\/strong>&nbsp;for Whisper to understand every sentence while its runtime collapsed from 25 to barely 8 minutes.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Why faster costs less<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Because billing is pegged to audio duration,&nbsp;<strong>a 3x-sped file burns roughly one-third of the tokens and one-third of the wall-clock time<\/strong>. In his test, charges fell from 0.24 USD to 0.07 USD at 2x and just 0.04 USD at 3x\u2014an instant 83 % saving.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">How to try it<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Download the talk with yt-dlp, then run&nbsp;<code>ffmpeg -i talk.mp3 -filter:a \"atempo=3.0\" talk_3x.mp3<\/code>.&nbsp;<strong>Upload the new file to ChatGPT and watch the transcript arrive in a fraction of the usual time<\/strong>. Going beyond 3x, however, degrades clarity and risks transcription errors.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>OpenAI\u2019s pay-as-you-go transcription looks straightforward: longer recordings mean more tokens and higher bills.&nbsp;Developer George Mandis stumbled on a way to shrink both time and cost dramatically&nbsp;by literally making the minutes shorter. A simple speed hack Mandis fed his audio through FFmpeg\u2019s atempo filter, tripling playback speed without pitch distortion.&nbsp;The accelerated track remained clear enough&nbsp;for Whisper &hellip; <a href=\"https:\/\/cms-articles.softonic.io\/en\/openai-charges-for-every-minute-its-ai-transcribes-so-they-have-found-a-way-to-fool-the-system\/\" class=\"more-link\">Continue reading<span class=\"screen-reader-text\"> &#8220;OpenAI charges for every minute its AI transcribes, so they have found a way to fool the system&#8221;<\/span><\/a><\/p>\n","protected":false},"author":9317,"featured_media":339290,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":"","wpcf-pageviews":0},"categories":[1015],"tags":[],"usertag":[],"vertical":[],"content-category":[],"class_list":["post-339289","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-news"],"aioseo_notices":[],"_links":{"self":[{"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/posts\/339289","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/users\/9317"}],"replies":[{"embeddable":true,"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/comments?post=339289"}],"version-history":[{"count":1,"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/posts\/339289\/revisions"}],"predecessor-version":[{"id":339291,"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/posts\/339289\/revisions\/339291"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/media\/339290"}],"wp:attachment":[{"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/media?parent=339289"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/categories?post=339289"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/tags?post=339289"},{"taxonomy":"usertag","embeddable":true,"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/usertag?post=339289"},{"taxonomy":"vertical","embeddable":true,"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/vertical?post=339289"},{"taxonomy":"content-category","embeddable":true,"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/content-category?post=339289"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}