---
title: "Upload an audio or video file and get a transcript"
description: "Submit several files at once on the Transcriptions page, set output formats and language, follow the processing status, and view or download the transcript"
canonical_url: "https://docs.ardge.com/en/knowledge/scribestt/upload-and-transcribe"
---
# Upload an audio or video file and get a transcript

> Submit several files at once on the Transcriptions page, set output formats and language, follow the processing status, and view or download the transcript

## Overview

One-off files — meeting recordings, interview footage — can be uploaded straight to ScribeSTT's **Transcriptions** page, with no connection or task to set up first. Speech recognition starts automatically after the upload, and the transcript is produced in the formats you chose.

This article covers creating a transcription on the **Transcriptions** page, following its processing status, and viewing and downloading the transcript.

## What you'll learn

- Submit several audio or video files for transcription at once
- Set the output formats, the audio's language, and the glossary to apply
- Follow a transcription's status in the list and the detail view, and retrieve the transcript files

**Estimated time: about 5 minutes**, not counting speech recognition itself, which depends on the audio's length

## Applies to

- ARX OS 1.0 and later
- ScribeSTT 1.3.0 and later

## Before you begin

- An administrator has set up the speech recognition model provider. Without it, transcription fails with "The speech recognition model isn't ready or isn't deployed. Confirm the model is running and try again." See [Set up the speech recognition model provider in ScribeSTT](https://docs.ardge.com/en/knowledge/scribestt/model-provider)

## Step 1: Open the create transcription dialog

1. In the **Workspace** group in the left navigation, go to the **Transcriptions** page
2. Click **New** at the top left of the page

The **Create new transcription** dialog opens, described as "Upload files to create a new transcription task".

![The Transcript page](https://docs.ardge.com/screenshots/knowledge/en/scribestt/upload-and-transcribe/step-1-1.webp)

![Create New Transcript](https://docs.ardge.com/screenshots/knowledge/en/scribestt/upload-and-transcribe/step-1-2.webp)

## Step 2: Select the files to transcribe

1. In the **Select files** field, click **Add files** and select your files, or drag them onto the dashed area below
2. Check the files and sizes in the list. Click the remove icon at the right of a row to drop that file

<note>

**Limits:**

Up to 20 files per submission, 1 GB per file

**Supported formats**: 12 audio and video extensions — see [What file formats and size limits does ScribeSTT support?](https://docs.ardge.com/en/knowledge/scribestt/formats-and-limits)

</note>

Files that don't qualify stay in the list outlined in red, with the reason below them:

<table>
<thead>
  <tr>
    <th>
      Message
    </th>
    
    <th>
      Reason
    </th>
  </tr>
</thead>

<tbody>
  <tr>
    <td>
      <code>
        File is empty
      </code>
    </td>
    
    <td>
      The file has no content
    </td>
  </tr>
  
  <tr>
    <td>
      <code>
        File too large (1 GB maximum)
      </code>
    </td>
    
    <td>
      The file exceeds 1 GB
    </td>
  </tr>
  
  <tr>
    <td>
      <code>
        Duplicate file
      </code>
    </td>
    
    <td>
      The same file is already in the list
    </td>
  </tr>
  
  <tr>
    <td>
      <code>
        Unsupported file format. Accepted: …
      </code>
    </td>
    
    <td>
      The extension isn't on the supported list
    </td>
  </tr>
  
  <tr>
    <td>
      <code>
        20 files maximum
      </code>
    </td>
    
    <td>
      More than 20 files were selected
    </td>
  </tr>
</tbody>
</table>

Once no rows are outlined in red, you can move on.

## Step 3: Set the output formats, language, and glossary

Fill in the rest of the dialog:

<table>
<thead>
  <tr>
    <th>
      Field
    </th>
    
    <th>
      Value
    </th>
    
    <th>
      Rules and notes
    </th>
  </tr>
</thead>

<tbody>
  <tr>
    <td>
      <strong>
        Output formats
      </strong>
    </td>
    
    <td>
      <strong>
        TXT
      </strong>
      
      , <strong>
        SRT
      </strong>
      
      , <strong>
        JSON
      </strong>
    </td>
    
    <td>
      Required, multiple allowed. With none selected it shows "Select at least one output format"
    </td>
  </tr>
  
  <tr>
    <td>
      <strong>
        Language
      </strong>
    </td>
    
    <td>
      The audio's language
    </td>
    
    <td>
      Optional. Chinese (<code>
        zh
      </code>
      
      ), English (<code>
        en
      </code>
      
      ), Japanese (<code>
        ja
      </code>
      
      ), or Korean (<code>
        ko
      </code>
      
      ); left empty, speech recognition detects it
    </td>
  </tr>
  
  <tr>
    <td>
      <strong>
        Glossary
      </strong>
    </td>
    
    <td>
      An existing glossary
    </td>
    
    <td>
      Optional. Names in the dropdown carry a <code>
        #
      </code>
      
       marking that glossary's language
    </td>
  </tr>
</tbody>
</table>

![Language options](https://docs.ardge.com/screenshots/knowledge/en/scribestt/upload-and-transcribe/step-3.webp)

<tip>

**You can't add output formats afterwards:**

The formats you select in the dialog decide which transcripts get produced. Formats you didn't select are never produced, and you can't switch to them later in **Transcription details**.

</tip>

## Step 4: Submit and follow the processing status

1. Click **Create**
2. Watch each record's **Status** column in the list

After submitting, the page shows "Created N/M transcription files". If some files failed to be created, it shows "Created N transcriptions, M failed: <file names="">

" instead.

</file>



The **Status** column moves through **Queued** and **Processing** to **Completed**. Three other statuses exist: **Failed**, **Canceled**, and **Skipped**.

Next to the status label, the **Status** column shows the actions available in that state: **Queued** and **Processing** can be canceled, **Failed** and **Canceled** can be retried. For what to do in each state, see [What to do when a transcription fails](https://docs.ardge.com/en/knowledge/scribestt/transcription-failures).

## Step 5: View and download the transcript

1. Click the file name in the list to open **Transcription details**
2. Switch between the **SRT**, **TXT**, and **JSON** tabs to see each format. Only the formats you selected in Step 3 appear as tabs
3. Click **Download (.)** at the top right to retrieve the current tab's file

**Transcription details** shows different content depending on the record's status:

<table>
<thead>
  <tr>
    <th>
      Status
    </th>
    
    <th>
      What you see
    </th>
  </tr>
</thead>

<tbody>
  <tr>
    <td>
      <strong>
        Queued
      </strong>
      
       / <strong>
        Processing
      </strong>
    </td>
    
    <td>
      "Transcription in progress……"
    </td>
  </tr>
  
  <tr>
    <td>
      <strong>
        Completed
      </strong>
    </td>
    
    <td>
      <strong>
        Processing time
      </strong>
      
      , the format tabs, and the transcript
    </td>
  </tr>
  
  <tr>
    <td>
      <strong>
        Failed
      </strong>
    </td>
    
    <td>
      "Transcription failed" and the failure reason
    </td>
  </tr>
  
  <tr>
    <td>
      <strong>
        Canceled
      </strong>
    </td>
    
    <td>
      "Transcription canceled"
    </td>
  </tr>
</tbody>
</table>

<tip>

**Turn on wrapping for long lines:**

The **Wrap** toggle at the bottom right of the content area wraps the transcript to the window width, so you don't have to scroll sideways.

</tip>

You can also retrieve several at once from the list: select the records to download, then click **Download (.)** in the toolbar. Click the arrow beside it to switch to `.txt`, `.srt`, or `.json`. When it's done, the page shows "Downloaded N transcription files".

![Transcript details, SRT](https://docs.ardge.com/screenshots/knowledge/en/scribestt/upload-and-transcribe/step-5-1.webp)

![Transcript details, TXT](https://docs.ardge.com/screenshots/knowledge/en/scribestt/upload-and-transcribe/step-5-2.webp)

## Finding a particular transcription in the list

The **Transcriptions** page lists only transcriptions created by upload. Transcriptions produced by a batch task aren't here — look in **Run details** on the **Tasks** page.

The list offers three ways to narrow it down:

- **Search box** — enter part of a file name
- **Filter › File type** — filter by audio extension
- **Filter › Transcription status** — filter by **Completed**, **Processing**, **Queued**, **Failed**, **Canceled**, or **Skipped**

The number of applied filters appears on the **Filter** button; **Reset** at the bottom of the menu clears them all at once.

## Next steps

When your files already sit on a NAS and you'd rather not upload them one by one, use batch transcription instead — see [Connect a NAS data source and create a batch transcription task](https://docs.ardge.com/en/knowledge/scribestt/nas-batch-transcription).

## Related articles

- [What file formats and size limits does ScribeSTT support?](https://docs.ardge.com/en/knowledge/scribestt/formats-and-limits)
- [Create a glossary to improve terminology accuracy](https://docs.ardge.com/en/knowledge/scribestt/glossary)
- [What to do when a transcription fails](https://docs.ardge.com/en/knowledge/scribestt/transcription-failures)


## Sitemap

See the full [sitemap](https://docs.ardge.com/sitemap.md) for all pages.