{"id":2741,"date":"2026-03-16T09:28:16","date_gmt":"2026-03-16T09:28:16","guid":{"rendered":"https:\/\/interactivehpc.dk\/?page_id=2741"},"modified":"2026-03-16T09:28:16","modified_gmt":"2026-03-16T09:28:16","slug":"transcriber-guide-2","status":"publish","type":"page","link":"https:\/\/interactivehpc.dk\/?page_id=2741&lang=da","title":{"rendered":"Transcriber Guide"},"content":{"rendered":"\n<h3 class=\"wp-block-heading gp-gutenbergpro-dab93\">What is Transcriber<\/h3>\n\n\n\n<p class=\"gp-gutenbergpro-dc0a9\">Transcriber is an application on <a href=\"https:\/\/cloud.sdu.dk\/app\/jobs\/create?app=transcriber-gui\">UCloud<\/a> designed to automatically convert audio and video files into accurate, readable text. It leverages advanced speech recognition models to transcribe spoken content, making it easier to analyze, search, and share information from interviews, lectures, meetings, podcasts, and other recordings. <\/p>\n\n\n\n<p class=\"gp-gutenbergpro-a80aa\">Transcriber helps researchers save time and effort by providing fast, reliable transcriptions directly within the secure UCloud platform &#8211; no technical expertise required. Whether you need simple text output or more advanced features like speaker identification and multiple file formats, Transcriber streamlines the process of turning your recordings into useful, accessible documents.<\/p>\n\n\n\n<h4 class=\"wp-block-heading gp-gutenbergpro-25fa9\">Which Transcriber should I use?<\/h4>\n\n\n\n<p class=\"gp-gutenbergpro-6e2b0\">There are two ways to use Transcriber on UCloud:<\/p>\n\n\n\n<figure class=\"wp-block-table\"><table class=\"has-fixed-layout\"><thead><tr><th>Interface (Default)<\/th><th>Batch<\/th><\/tr><\/thead><tbody><tr><td>Simple, modern screen with buttons<\/td><td>No screen, just settings interface<\/td><\/tr><tr><td>Designed for most users<\/td><td>Designed for advanced users and batch jobs<\/td><\/tr><tr><td>Add files by drag and drop from computer or UCloud folders<\/td><td>Add files from UCloud folders only<\/td><\/tr><tr><td>Download results from the app or UCloud folders<\/td><td>Download results from UCloud folders<\/td><\/tr><tr><td>Basic settings: language and model selection<\/td><td>Advanced settings: language, model, number of speakers, and merge speaker entries<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<p class=\"gp-gutenbergpro-d833a\">Select the guide of your choice below for step-by-step instructions for each version.<\/p>\n\n\n\n<p class=\"gp-gutenbergpro-5f2f0\">=== &#8220;Transcriber Interface (Default version)&#8221;<\/p>\n\n\n\n<p class=\"gp-gutenbergpro-08e02\">This guide will walk you through using the latest <strong>default<\/strong> version of Transcriber, featuring an improved, user-friendly interface for transcribing your audio and video files on UCloud.<\/p>\n\n\n\n<h3 class=\"wp-block-heading gp-gutenbergpro-d3a1f\">1. Using the Transcriber default (Interface) application<\/h3>\n\n\n\n<h4 class=\"wp-block-heading gp-gutenbergpro-7a6ea\">1.1 Finding and launching the application<\/h4>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Go to the UCloud application page<\/strong> and use the search function to find <strong>Transcriber<\/strong>.<br><figure><img decoding=\"async\" src=\"\/wp-content\/uploads\/2026\/03\/Transcriberguide16.jpg\" alt=\"Transcriber Guide Screen\"><\/figure><\/li>\n\n\n\n<li><strong>Open the app<\/strong> by clicking on <strong>Transcriber<\/strong>.<br><figure><img decoding=\"async\" src=\"\/wp-content\/uploads\/2026\/03\/Transcriberguide17.jpg\" alt=\"Transcriber Node Assigned\"><\/figure><\/li>\n<\/ul>\n\n\n\n<h4 class=\"wp-block-heading gp-gutenbergpro-3a30d\">1.2 Configuring your job<\/h4>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Name your job<\/strong>: Choose a name that helps you identify the job later (e.g., &#8220;Transcriber Demo 1&#8221;).\n<ul class=\"wp-block-list\">\n<li><em>Note<\/em>: Avoid special characters like &#8220;\u00e6\u00f8\u00e5&#8221;.<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>Set the duration<\/strong>: Specify how many hours the job should run. For reference, a 1-hour audio typically takes about 1 hour to transcribe on a <code>u1-standard-16<\/code> machine.\n<ul class=\"wp-block-list\">\n<li>You can stop the machine early or add more time later if needed.<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>Choose a machine<\/strong>: We recommend <span class=\"machine-name\">u3-gpu-1<\/span> if available, otherwise use <span class=\"machine-name\">u1-standard-16<\/span>.<\/li>\n\n\n\n<li><strong>(Optional) Select folders to use<\/strong>: If you want to use UCloud folders, select the folder(s) containing your files.\n<ul class=\"wp-block-list\">\n<li>The app will scan these folders for compatible files.<br><br><blockquote><p><strong>Note:<\/strong> Folders named <code>UPLOADS<\/code> and <code>COMPLETED<\/code> are reserved by the app and won&#8217;t be scanned.<\/p><\/blockquote><\/li>\n<\/ul>\n<\/li>\n\n\n\n<li>Once you finish configuring, click <strong>Submit<\/strong> to start the Transcriber job.<\/li>\n<\/ul>\n\n\n\n<figure class=\"wp-block-image\"><img loading=\"lazy\" decoding=\"async\" width=\"1221\" height=\"838\" src=\"https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/TranscriberguideGUI1-1.jpg\" alt=\"Starting the job on UCloud\" class=\"wp-image-2619\" srcset=\"https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/TranscriberguideGUI1-1.jpg 1221w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/TranscriberguideGUI1-1-300x206.jpg 300w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/TranscriberguideGUI1-1-1024x703.jpg 1024w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/TranscriberguideGUI1-1-768x527.jpg 768w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/TranscriberguideGUI1-1-1200x824.jpg 1200w\" sizes=\"auto, (max-width: 1221px) 100vw, 1221px\" \/><\/figure>\n\n\n\n<p class=\"gp-gutenbergpro-82eb7\">After submitting your job, you&#8217;ll be redirected to a new page where your Transcriber machine is being prepared. Once it&#8217;s ready, click the <strong>Open interface<\/strong> button to launch the Transcriber application.<\/p>\n\n\n\n<figure class=\"wp-block-image\"><img loading=\"lazy\" decoding=\"async\" width=\"1144\" height=\"1013\" src=\"https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/TranscriberguideGUI2-1.jpg\" alt=\"Open interface\" class=\"wp-image-2621\" srcset=\"https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/TranscriberguideGUI2-1.jpg 1144w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/TranscriberguideGUI2-1-300x266.jpg 300w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/TranscriberguideGUI2-1-1024x907.jpg 1024w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/TranscriberguideGUI2-1-768x680.jpg 768w\" sizes=\"auto, (max-width: 1144px) 100vw, 1144px\" \/><\/figure>\n\n\n\n<h4 class=\"wp-block-heading gp-gutenbergpro-bbfe6\">1.3 Adding files to be transcribed<\/h4>\n\n\n\n<p class=\"gp-gutenbergpro-f66e7\">Once the app starts, decide how you want to add files for transcription. You have two main options:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Use files from UCloud folders<\/strong>\n<ul class=\"wp-block-list\">\n<li>If you <strong>selected folders<\/strong> in the launch step, the app will automatically list the files detected there.<\/li>\n\n\n\n<li><strong>Choose<\/strong> the files you want and click <strong>Add UCloud files<\/strong> to add them to your transcription queue.<\/li>\n\n\n\n<li>If you add new files to the folder after the job has started, click <strong>Scan UCloud folder<\/strong> to refresh the list.<br><blockquote><p><strong>Note:<\/strong> This section is only visible if you selected a folder when launching your Transcriber job on UCloud.<\/p><\/blockquote><\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>Upload files from your computer<\/strong>\n<ul class=\"wp-block-list\">\n<li><strong>Drag and Drop<\/strong> your files directly into the upload area of the Transcriber app.<\/li>\n\n\n\n<li>Or <strong>click<\/strong> the upload area to browse your computer and select files.<\/li>\n\n\n\n<li>Multiple files can be added at once.<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p class=\"gp-gutenbergpro-3e151\"><strong>Info:<\/strong> The app can only process <code>.mp3<\/code>, <code>.mp4<\/code>, <code>.m4a<\/code>, <code>.wav<\/code>, and <code>.mpg<\/code> files. If your file is in another format, we recommend using VLC to convert it. VLC can be downloaded from the Software Center\/Company Portal.<\/p>\n<\/blockquote>\n\n\n\n<figure class=\"wp-block-image\"><img loading=\"lazy\" decoding=\"async\" width=\"1081\" height=\"1097\" src=\"https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/TranscriberguideGUI3-1.jpg\" alt=\"Selecting files to be transcribed\" class=\"wp-image-2623\" srcset=\"https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/TranscriberguideGUI3-1.jpg 1081w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/TranscriberguideGUI3-1-296x300.jpg 296w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/TranscriberguideGUI3-1-1009x1024.jpg 1009w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/TranscriberguideGUI3-1-768x779.jpg 768w\" sizes=\"auto, (max-width: 1081px) 100vw, 1081px\" \/><\/figure>\n\n\n\n<h4 class=\"wp-block-heading gp-gutenbergpro-94487\">1.4 Starting the transcription<\/h4>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Once you have added all your desired files to the transcription queue, click <strong>Start Transcription<\/strong>.<\/li>\n\n\n\n<li>The app will begin transcribing your files and show a progress bar so you can track the transcription status in real time.<br><br><blockquote><p><strong>Note:<\/strong> The progress bar provides an estimated completion time for each file, but this estimate may change as the transcription proceeds. Factors such as the selected machine, the amount of speech in the audio, and the selected transcription model can affect how long each file takes.<\/p><\/blockquote><\/li>\n<\/ul>\n\n\n\n<h4 class=\"wp-block-heading gp-gutenbergpro-26ea1\">1.5 Downloading your transcriptions<\/h4>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>While the job is running<\/strong>: You can download completed transcriptions directly from the app, either one by one or as a zip file. When downloading individual files, you can select your preferred output format (TXT, DOCX, VTT, etc.). If you choose to download as a zip file, you&#8217;ll receive all available output formats for each transcription.<br><ul><br><li><strong>Important:<\/strong> If you are working with confidential or sensitive data, remember that you may <strong>not<\/strong> download it directly to any personal or nonapproved devices.<\/li><\/ul><br><blockquote><p><strong>Note:<\/strong> For better readability and to save time on post-processing, you can download a merged speaker format of the transcription that combines consecutive text entries from the same speaker into natural, flowing sentences. This feature helps streamline your workflow by reducing the need for manual text editing.<\/p><\/blockquote><\/li>\n<\/ul>\n\n\n\n<figure class=\"wp-block-image\"><img loading=\"lazy\" decoding=\"async\" width=\"915\" height=\"1257\" src=\"https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/TranscriberguideGUI4-1.jpg\" alt=\"Download transcribed files\" class=\"wp-image-2624\" srcset=\"https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/TranscriberguideGUI4-1.jpg 915w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/TranscriberguideGUI4-1-218x300.jpg 218w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/TranscriberguideGUI4-1-745x1024.jpg 745w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/TranscriberguideGUI4-1-768x1055.jpg 768w\" sizes=\"auto, (max-width: 915px) 100vw, 915px\" \/><\/figure>\n\n\n\n<p class=\"gp-gutenbergpro-950d1\"><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>After the job is finished<\/strong>: All transcriptions will be available on UCloud in the folder: <code>\/Jobs\/Transcriber\/&lt;job-id&gt;\/TRANSCRIPTIONS\/<\/code>.<\/li>\n<\/ul>\n\n\n\n<h4 class=\"wp-block-heading gp-gutenbergpro-fbf86\">1.6 Optional: Adjusting settings<\/h4>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Click <strong>Show settings<\/strong> at the top of the page to adjust:\n<ul class=\"wp-block-list\">\n<li><strong>New feature:<\/strong> Transcribe and stop This setting allows the job to stop automatically once all selected files have been transcribed. Enabling this option is highly efficient in terms of compute hours allocated to your project, as the job remains active only while work is being performed. As always, the transcribed files can be found in the UCloud filesystem, either in the job folder or in the folder selected at startup.<\/li>\n\n\n\n<li>The transcription model (default is &#8220;large-v3&#8221;).<\/li>\n\n\n\n<li>The language (default is &#8220;Automatic&#8221;).<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p class=\"gp-gutenbergpro-def07\"><strong>Note:<\/strong> If you&#8217;re unsure, the default settings are usually best.<\/p>\n<\/blockquote>\n\n\n\n<p class=\"gp-gutenbergpro-db4ab\"><strong>Need more advanced options?<\/strong><br>Try the Transcriber batch version, which offers extended configuration possibilities.<\/p>\n\n\n\n<p class=\"gp-gutenbergpro-497dd\">=== &#8220;Transcriber Batch&#8221;<\/p>\n\n\n\n<p class=\"gp-gutenbergpro-ac694\">This guide provides step-by-step instructions for using the <strong>Transcriber Batch<\/strong> application, which offers advanced configuration options and is optimized for efficient, large-scale transcription tasks.<\/p>\n\n\n\n<h3 class=\"wp-block-heading gp-gutenbergpro-af171\">1. Using the Transcriber batch application<\/h3>\n\n\n\n<h4 class=\"wp-block-heading gp-gutenbergpro-89a3a\">1.1 Finding the application<\/h4>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Go into the application and use the search function to find <strong>Transcriber<\/strong>.<figure><img decoding=\"async\" src=\"\/wp-content\/uploads\/2026\/03\/Transcriberguide16.jpg\" alt=\"Transcriber Guide Screen\"><\/figure><\/li>\n\n\n\n<li>Click on <strong>Transcriber<\/strong>.<figure><img decoding=\"async\" src=\"\/wp-content\/uploads\/2026\/03\/Transcriberguide17.jpg\" alt=\"Transcriber Node Assigned\"><\/figure><\/li>\n<\/ul>\n\n\n\n<h4 class=\"wp-block-heading gp-gutenbergpro-ed97f\">1.2 Using the application<\/h4>\n\n\n\n<p class=\"gp-gutenbergpro-92c42\">You should now see the following screen:<\/p>\n\n\n\n<figure class=\"wp-block-image\"><img loading=\"lazy\" decoding=\"async\" width=\"1057\" height=\"1113\" src=\"https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide18-1.jpg\" alt=\"Transcriber Node Assigned\" class=\"wp-image-2625\" srcset=\"https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide18-1.jpg 1057w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide18-1-285x300.jpg 285w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide18-1-972x1024.jpg 972w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide18-1-768x809.jpg 768w\" sizes=\"auto, (max-width: 1057px) 100vw, 1057px\" \/><\/figure>\n\n\n\n<p class=\"gp-gutenbergpro-d3418\"><\/p>\n\n\n\n<p class=\"gp-gutenbergpro-de290\">There are several options here, and it can seem overwhelming. For this example, we&#8217;ll walk through the quickest way to start a transcription.<\/p>\n\n\n\n<h5 class=\"wp-block-heading gp-gutenbergpro-90a0f\">1.2.1 Choose a name for your job<\/h5>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Pick a name that makes it easy to find your data later and distinguish between different jobs.<br><ul><br><li>Example: &#8220;Transcriber demo 1&#8221;.<\/li><\/ul><br><blockquote><p><strong>Note<\/strong>: Job and file names cannot include special characters such as &#8220;\u00e6\u00f8\u00e5&#8221;.<\/p><\/blockquote><\/li>\n<\/ul>\n\n\n\n<h5 class=\"wp-block-heading gp-gutenbergpro-089b6\">1.2.2 Select the duration of your job<\/h5>\n\n\n\n<ul class=\"wp-block-list\">\n<li>The application can transcribe in 1:1 time. For instance, a 1-hour audio file will take approximately 1 hour to transcribe.<\/li>\n\n\n\n<li>We recommend allocating <strong>double<\/strong> the length of the audio file to avoid interruptions.<br><ul><br><li>Example: For a 1-hour audio file, allocate 2 hours.<\/li><\/ul><br><blockquote><p><strong>Note<\/strong>: If you run out of allocated time, the file being transcribed will fail. You can allocate more time after starting the job if needed.<\/p><\/blockquote><\/li>\n<\/ul>\n\n\n\n<h5 class=\"wp-block-heading gp-gutenbergpro-fc4f7\">1.2.3 Pick a machine to use<\/h5>\n\n\n\n<ul class=\"wp-block-list\">\n<li>We recommend the <span class=\"machine-name\">u3-gpu-1<\/span> machine, which performed best in our tests. If the option is unavailable we recommend the <span class=\"machine-name\">u1-standard-16<\/span> as an alternative.<br><br><blockquote><p>Feel free to test with sample files to see what works best for you.<\/p><\/blockquote><br><figure><img decoding=\"async\" src=\"\/wp-content\/uploads\/2026\/03\/Transcriberguide19.jpg\" alt=\"Transcriber Node Assigned\"><\/figure><br><\/li>\n<\/ul>\n\n\n\n<h5 class=\"wp-block-heading gp-gutenbergpro-3ffd9\">1.2.4 Select the input file<\/h5>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Click the &#8220;use&#8221; button.<figure><img decoding=\"async\" src=\"\/wp-content\/uploads\/2026\/03\/Transcriberguide20.jpg\" alt=\"Transcriber Node Assigned\"><\/figure><\/li>\n\n\n\n<li>Click the text box to select your file.<figure><img decoding=\"async\" src=\"\/wp-content\/uploads\/2026\/03\/Transcriberguide21.jpg\" alt=\"Transcriber Node Assigned\"><\/figure><\/li>\n\n\n\n<li>Navigate to your &#8220;drives&#8221; and select the folder with your file or click &#8220;use&#8221; if it&#8217;s already listed.<figure><img decoding=\"async\" src=\"\/wp-content\/uploads\/2026\/03\/Transcriberguide22.jpg\" alt=\"Transcriber Node Assigned\"><\/figure><\/li>\n<\/ul>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p class=\"gp-gutenbergpro-1a169\"><strong>Note<\/strong>: The app can only process <code>.mp3<\/code>, <code>.mp4<\/code>, <code>.m4a<\/code>, <code>.wav<\/code>, and <code>.mpg<\/code> files. If your file is in another format, we recommend using VLC to convert it. VLC can be downloaded from the Software Center\/Company Portal.<\/p>\n<\/blockquote>\n\n\n\n<h5 class=\"wp-block-heading gp-gutenbergpro-dbd0f\">1.2.5 Select the output directory<\/h5>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Choose where your output will be saved.<\/li>\n\n\n\n<li>Click &#8220;use&#8221; on &#8220;option: &#8211;output_dir&#8221;.<figure><img decoding=\"async\" src=\"\/wp-content\/uploads\/2026\/03\/Transcriberguide23.jpg\" alt=\"Transcriber Node Assigned\"><\/figure><\/li>\n\n\n\n<li>Select the folder you want for your transcription output.<figure><img decoding=\"async\" src=\"\/wp-content\/uploads\/2026\/03\/Transcriberguide24.jpg\" alt=\"Transcriber Node Assigned\"><\/figure><\/li>\n<\/ul>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p class=\"gp-gutenbergpro-271e7\"><strong>Note<\/strong>: The app supports <code>.mp3<\/code>, <code>.mp4<\/code>, <code>.m4a<\/code>, <code>.wav<\/code>, and <code>.mpg<\/code> files. For other formats, consider converting using VLC.<\/p>\n<\/blockquote>\n\n\n\n<p class=\"gp-gutenbergpro-785b0\">Now, you are ready to begin your transcription. Click <strong>Submit<\/strong> to start the process.<\/p>\n\n\n\n<p class=\"gp-gutenbergpro-8f1e1\"><em><strong>There are additional options available. These are covered in the &#8220;Other Options&#8221; section.<\/strong><\/em><\/p>\n\n\n\n<figure class=\"wp-block-image\"><img loading=\"lazy\" decoding=\"async\" width=\"2002\" height=\"1114\" src=\"https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide25-1.jpg\" alt=\"Transcriber Node Assigned\" class=\"wp-image-2620\" srcset=\"https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide25-1.jpg 2002w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide25-1-300x167.jpg 300w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide25-1-1024x570.jpg 1024w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide25-1-768x427.jpg 768w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide25-1-1536x855.jpg 1536w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide25-1-1200x668.jpg 1200w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide25-1-1980x1102.jpg 1980w\" sizes=\"auto, (max-width: 2002px) 100vw, 2002px\" \/><\/figure>\n\n\n\n<p class=\"gp-gutenbergpro-977a9\">Once the process starts, you can close your computer. If you want to ensure everything is running smoothly, wait until a &#8220;node&#8221; is assigned. Your screen will look like this:<\/p>\n\n\n\n<figure class=\"wp-block-image\"><img loading=\"lazy\" decoding=\"async\" width=\"2032\" height=\"986\" src=\"https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide26-1.jpg\" alt=\"Transcriber Node Assigned\" class=\"wp-image-2617\" srcset=\"https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide26-1.jpg 2032w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide26-1-300x146.jpg 300w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide26-1-1024x497.jpg 1024w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide26-1-768x373.jpg 768w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide26-1-1536x745.jpg 1536w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide26-1-1200x582.jpg 1200w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide26-1-1980x961.jpg 1980w\" sizes=\"auto, (max-width: 2032px) 100vw, 2032px\" \/><\/figure>\n\n\n\n<p class=\"gp-gutenbergpro-01d4a\">Once the transcription is complete, you will see the following screen:<\/p>\n\n\n\n<figure class=\"wp-block-image\"><img loading=\"lazy\" decoding=\"async\" width=\"1436\" height=\"795\" src=\"https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide27-1.jpg\" alt=\"Transcriber Node Assigned\" class=\"wp-image-2618\" srcset=\"https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide27-1.jpg 1436w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide27-1-300x166.jpg 300w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide27-1-1024x567.jpg 1024w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide27-1-768x425.jpg 768w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide27-1-1200x664.jpg 1200w\" sizes=\"auto, (max-width: 1436px) 100vw, 1436px\" \/><\/figure>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p class=\"gp-gutenbergpro-3bee8\"><strong>Note<\/strong>: This is not the actual output of your transcription. The transcription files are located in the folder you selected for output. You\u2019ll find something like this:<\/p>\n<\/blockquote>\n\n\n\n<figure class=\"wp-block-image\"><img loading=\"lazy\" decoding=\"async\" width=\"2046\" height=\"926\" src=\"https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide28-1.jpg\" alt=\"Transcriber Node Assigned\" class=\"wp-image-2622\" srcset=\"https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide28-1.jpg 2046w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide28-1-300x136.jpg 300w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide28-1-1024x463.jpg 1024w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide28-1-768x348.jpg 768w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide28-1-1536x695.jpg 1536w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide28-1-1200x543.jpg 1200w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide28-1-1980x896.jpg 1980w\" sizes=\"auto, (max-width: 2046px) 100vw, 2046px\" \/><\/figure>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<h4 class=\"wp-block-heading gp-gutenbergpro-32738\">1.3 Transcription output formats<\/h4>\n\n\n\n<p class=\"gp-gutenbergpro-0ab1c\">You will have several different files with your transcription. Commonly used formats include <code>.txt<\/code> and <code>.docx<\/code>. You can choose the format that suits your needs best. If you want a specific output format, refer to the &#8220;output_format&#8221; section under &#8220;Other Options&#8221;.<\/p>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p class=\"gp-gutenbergpro-8e847\"><strong>Note<\/strong>: It is not possible to preview <code>.docx<\/code> files directly within UCloud. To view the transcription in <code>.docx<\/code> format, you must first download the file to your own computer and open it with Microsoft Word or another compatible program.<\/p>\n<\/blockquote>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<h3 class=\"wp-block-heading gp-gutenbergpro-ba8bd\">2. Optional parameters<\/h3>\n\n\n\n<figure class=\"wp-block-image\"><img loading=\"lazy\" decoding=\"async\" width=\"1252\" height=\"546\" src=\"https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide32-1.jpg\" alt=\"Transcriber Optional Parameters\" class=\"wp-image-2616\" srcset=\"https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide32-1.jpg 1252w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide32-1-300x131.jpg 300w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide32-1-1024x447.jpg 1024w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide32-1-768x335.jpg 768w, https:\/\/interactivehpc.dk\/wp-content\/uploads\/2026\/03\/Transcriberguide32-1-1200x523.jpg 1200w\" sizes=\"auto, (max-width: 1252px) 100vw, 1252px\" \/><\/figure>\n\n\n\n<h4 class=\"wp-block-heading gp-gutenbergpro-951a0\">2.1 Option: &#8211;output_format<\/h4>\n\n\n\n<p class=\"gp-gutenbergpro-5108a\">By default, the application produces all 8 formats. You can limit the output to a specific format by selecting one of the following:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>CSV<\/strong>: Contains all parameters outputted from the Whisper model.<\/li>\n\n\n\n<li><strong>SRT<\/strong>: SubRip file format, a widely adopted subtitle format.<\/li>\n\n\n\n<li><strong>TXT<\/strong>: Pure text file with the transcription.<\/li>\n\n\n\n<li><strong>VTT<\/strong>: Web Video Text Tracks format, includes timestamps.<\/li>\n\n\n\n<li><strong>JSON<\/strong>: JavaScript Object Notation.<\/li>\n\n\n\n<li><strong>TSV<\/strong>: Tab-separated values file containing start, end, and text.<\/li>\n\n\n\n<li><strong>DOTE<\/strong>: Transcription software developed by the BigSoftVideo team at AAU.<\/li>\n\n\n\n<li><strong>DOCX<\/strong>: Text file with transcription and speaker recognition.<\/li>\n<\/ul>\n\n\n\n<h4 class=\"wp-block-heading gp-gutenbergpro-216b7\">2.2 Option: &#8211;output_model<\/h4>\n\n\n\n<p class=\"gp-gutenbergpro-fb8f5\">Select the model size:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Small<\/strong>: Faster but less accurate.<\/li>\n\n\n\n<li><strong>Medium<\/strong>: Slightly slower, more accurate.<\/li>\n\n\n\n<li><strong>Large<\/strong>: Most accurate but slowest.<\/li>\n<\/ul>\n\n\n\n<p class=\"gp-gutenbergpro-61297\">The default is <strong>Large<\/strong>. With a machine featuring 16 vCPUs and 96GB of memory, transcription speed is about the same as the audio length (e.g., 1 minute of audio takes approximately 1 minute to transcribe).<\/p>\n\n\n\n<h4 class=\"wp-block-heading gp-gutenbergpro-025ed\">2.3 Option: &#8211;output_language<\/h4>\n\n\n\n<p class=\"gp-gutenbergpro-aa10c\">Specify the language for transcription. The Whisper model can detect and automatically choose the language. If you select a language manually, the model will translate audio into that language.<\/p>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p class=\"gp-gutenbergpro-b0c31\"><strong>Note<\/strong>: The detected or chosen language determines the output language. For example, if the chosen language is English, the model will translate multiple languages into English.<\/p>\n<\/blockquote>\n\n\n\n<h4 class=\"wp-block-heading gp-gutenbergpro-79f1a\">2.4 Interactive mode<\/h4>\n\n\n\n<p class=\"gp-gutenbergpro-6935a\">Enable interactive mode for access to the app terminal or a web interface. The web interface includes a JupyterLab workspace for working with notebooks.<\/p>\n\n\n\n<h4 class=\"wp-block-heading gp-gutenbergpro-734c8\">2.5 Archive password<\/h4>\n\n\n\n<p class=\"gp-gutenbergpro-b31a9\">Encrypt and password-protect the ZIP output archive. Specify a password for the archive as a text string.<\/p>\n\n\n\n<h4 class=\"wp-block-heading gp-gutenbergpro-73610\">2.6 Minimum and maximum number of speakers<\/h4>\n\n\n\n<p class=\"gp-gutenbergpro-5d1f5\">Specify the number of speakers to improve speaker diarization accuracy in some cases.<\/p>\n\n\n\n<h4 class=\"wp-block-heading gp-gutenbergpro-8d1d1\">2.7 Merge consecutive text entries from the same speaker (Recommended)<\/h4>\n\n\n\n<p class=\"gp-gutenbergpro-3224e\">This option combines consecutive text entries from the same speaker into a single block, improving readability.<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>When enabled, the app generates additional files with merged text in <strong>docx<\/strong>, <strong>dote<\/strong>, <strong>json<\/strong>, and <strong>csv<\/strong> formats. These files are named <code>filename_merged<\/code> and are created alongside the original files.<br><br><blockquote><p>To make the option visible, scroll down in the optional parameter window.<\/p><\/blockquote><\/li>\n<\/ul>\n\n\n\n<h5 class=\"wp-block-heading gp-gutenbergpro-60377\">New feature: Transcribe and stop<\/h5>\n\n\n\n<p class=\"gp-gutenbergpro-4df8a\">Transcribe and stop is a new option in the Transcriber interface application that automatically stops the job once all selected files have been transcribed. Enabling this setting saves compute resources by ensuring the job only runs while actual transcription is taking place &#8211; making your allocation more efficient and freeing up resources so you can complete more transcriptions overall. <\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>How it works: When this feature is enabled, the job will finish and automatically shut down as soon as all transcriptions are complete.<\/li>\n\n\n\n<li>Access your results: Your transcribed files are always available in the UCloud filesystem, either in the job folder or in your chosen output folder at startup.<\/li>\n<\/ul>\n\n\n\n<h4 class=\"wp-block-heading gp-gutenbergpro-2fc0e\">Transcriber webinar<\/h4>\n\n\n\n<p class=\"gp-gutenbergpro-60567\">A recorded Transcriber webinar is available <a href=\"https:\/\/interactivehpc.dk\/?p=2203\" data-type=\"post\" data-id=\"2203\">here<\/a>.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>What is Transcriber Transcriber is an application on UCloud designed to automatically convert audio and video files into accurate, readable text. It leverages advanced speech recognition models to transcribe spoken content, making it easier to analyze, search, and share information from interviews, lectures, meetings, podcasts, and other recordings. Transcriber helps researchers save time and effort [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"gtp_columnspro_styling":"{}","gtp_paragraph_styling":"{\"6056788c-5072-4a2f-a6a4-6f1aa23f225c\":\" .gp-gutenbergpro-60567 { background-position-x: 50%;\\nbackground-position-y: 50%;\\nbackground-size: cover;\\nheight: px; }\"}","gtp_heading_styling":"{\"2fc0e336-2dcb-4da6-9ab1-464c1a91773d\":\" .gp-gutenbergpro-2fc0e { background-position-x: 50%;\\nbackground-position-y: 50%;\\nbackground-size: cover;\\nletter-spacing: 0px !important;\\ntext-shadow: 0px 0px 0px ;\\nfont-family:  !important; }\"}","gtp_spacer_styling":"{}","gtp_video_styling":"{}","gtp_group_styling":"{}","gtp_cover_styling":"{}","footnotes":""},"class_list":["post-2741","page","type-page","status-publish","hentry"],"lang":"da","translations":{"da":2741,"en":2583},"pll_sync_post":[],"_links":{"self":[{"href":"https:\/\/interactivehpc.dk\/index.php?rest_route=\/wp\/v2\/pages\/2741","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/interactivehpc.dk\/index.php?rest_route=\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/interactivehpc.dk\/index.php?rest_route=\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/interactivehpc.dk\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/interactivehpc.dk\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=2741"}],"version-history":[{"count":1,"href":"https:\/\/interactivehpc.dk\/index.php?rest_route=\/wp\/v2\/pages\/2741\/revisions"}],"predecessor-version":[{"id":2743,"href":"https:\/\/interactivehpc.dk\/index.php?rest_route=\/wp\/v2\/pages\/2741\/revisions\/2743"}],"wp:attachment":[{"href":"https:\/\/interactivehpc.dk\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=2741"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}