Skip to main content

Automatic Transcription

This guide explains what happens during automatic transcription processing.

What is Automatic Transcription?​

Automatic transcription uses speech recognition to convert audio into text. After upload, processing happens automatically on the server.

How It Works​

  1. Upload Completes - Your file finishes uploading to Scriptix
  2. Session Created - A transcription session starts with your settings
  3. Processing Begins - The server transcribes your audio automatically
  4. Transcript Ready - You can open and edit your completed transcript

The entire process runs in the background. You can close your browser and check back later.

Monitoring Your Transcription​

Finding Your Sessions​

  1. Click Workspace in the sidebar
  2. Your transcriptions appear in the list
  3. Look for your filename

What You'll See​

Each transcription shows:

  • Filename - The name of your file
  • Status - Current processing status (more on this below)
  • Duration - Length of your audio
  • Language - The language you selected
  • Date - When you uploaded it

The workspace helps you manage multiple transcriptions:

  • Pages - Navigate through pages if you have many files
  • Search - Find specific files by name
  • Sort - Order by date, name, or other fields
  • Folders - Organize transcriptions into folders

Understanding Status​

Your transcription goes through several stages:

StatusWhat It Means
UploadingFile is being uploaded to the server
QueuedWaiting to be processed
ProcessingCurrently being transcribed
FinishedReady to view and edit
FailedSomething went wrong

When It's Finished​

When status shows "Finished":

  • Click on the transcription to open it
  • You'll see the complete transcript with timestamps
  • Start editing right away

If It Fails​

If status shows "Failed":

  • You'll see an error message
  • Check the error for details
  • Try uploading again with different settings if needed

Transcription Settings​

Language​

The language setting determines which speech recognition model is used. Always select the exact language spoken in your audio for best results.

Speaker Diarization​

Identifies different speakers and labels them automatically.

When to use it:

  • ✅ Interviews with multiple people
  • ✅ Meetings and group discussions
  • ✅ Podcasts with co-hosts
  • ✅ Panel conversations

When to skip it:

  • ❌ Single person speaking (like lectures)
  • ❌ Narration or voice-overs
  • ❌ When you don't care who said what

Result: Speakers labeled as "Speaker 1", "Speaker 2", etc. You can rename them in the editor.

Folder Organization​

Automatically places transcriptions in a specific folder to keep your workspace organized.

What You Get​

Complete Transcript​

  • Every word from your audio typed out
  • Organized into paragraphs
  • Timestamps showing when each part was spoken

Speaker Labels (if enabled)​

  • Each speaker identified
  • Their parts labeled and separated
  • Easy to see who said what

Editable Document​

  • Click to open in the transcript editor
  • Edit any mistakes
  • Add notes or comments
  • Export in various formats

Checking Progress​

During Processing​

While your file is being transcribed:

  • You can close your browser - processing continues
  • Come back later to check the status
  • You'll find it in the workspace when ready

How Long Does It Take?​

Processing time depends on:

  • File length - Longer files take more time
  • Audio quality - Clear audio processes faster
  • Diarization - Adds extra time for speaker detection
  • Server load - Busy times may take longer

Generally: Processing often takes a fraction of your audio length, but can vary.

Getting Notified​

Currently:

  • Check your workspace to see when it's done
  • Status updates automatically when you refresh
  • Error messages appear if something goes wrong

Tips for Better Results​

Before Upload​

✅ Double-check language - This is the most important setting ✅ Enable diarization - If you have multiple speakers ✅ Choose a folder - Keep things organized from the start ✅ Test with a sample - Try a short clip first if unsure

While Processing​

✅ Be patient - Good transcription takes time ✅ Don't re-upload - Wait for the first attempt to complete ✅ Check for errors - If it fails, read the error message

After Completion​

✅ Review the transcript - Check overall quality ✅ Edit mistakes - Fix any errors you find ✅ Save your work - Changes save automatically in the editor

Next Steps​

Related Documentation: