Guide Transcribe

Transcription

Transcribe turns spoken audio into a downloadable file in your library. Upload an audio or video file, or paste a YouTube URL, and Scholarly keeps working in the background after you close the creator or leave the app.

Where to Start

Pick Transcribe from the tool row on Home. Then choose one source:

  • Upload an audio or video file.
  • Paste a YouTube URL for a video with an available transcript.

Each source can be up to 60 minutes long.

Choose the Output

The customization step offers two formats:

  • Plain text (.txt) -- clean transcript text for reading, search, notes, and AI tools.
  • Timestamped captions (.vtt) -- standard WebVTT cues for video players and caption editors.

You can also set an optional file name. Scholarly adds “transcript” and the selected extension for you.

Cost and Background Processing

One transcription uses 1 AI credit for the whole source, whether it is a short clip or the full 60 minutes. The credit is reserved when the task is queued and committed after the transcript file is created. If the task fails without producing a file, the credit is returned automatically.

The task appears in the background tasks strip on Home. You can close the creator, navigate elsewhere, or leave the app; processing continues on Scholarly's servers. When it finishes, the task opens the new transcript file, where you can preview, edit, download, chat with, or reuse it.

Tips and Limits

  • A YouTube video needs available captions or a transcript; live streams and unavailable videos cannot be transcribed.
  • Very quiet or silent media may fail because there are no spoken words to write.
  • For a transcript you want to read or edit, choose .txt. For timed captions, choose .vtt.
  • Retry is available for failed tasks. A retry reserves one credit again and returns it if that attempt fails.
Was this helpful?
PreviousText-to-SpeechNextAccount & Login