Uploading audio files

Get clean transcripts from lectures, recordings, and homilies by following these audio-prep tips.

May 21, 2026

Vulgate transcribes audio uploads into searchable, citable text. Lectures, sermons, conference talks, podcasts, and field recordings all work — the quality of the transcript depends almost entirely on the quality of the source audio.

Upload an audio file

  1. Open Uploads → Upload in the sidebar.
  2. Drag-and-drop the audio file into the upload area or pick it from disk.
  3. Make sure Transcription is checked (and add Music analysis if it's a musical recording).
  4. Submit to start the upload.

Audio is uploaded once, then queued for transcription. You can leave the page and come back; we'll notify you when it's ready.

Supported formats

MP3, M4A / AAC, WAV, FLAC, OGG, and Opus are all accepted, plus common video containers (.mp4, .mov) where the audio track is transcribed. See Supported file formats for the full table.

Tips for great transcripts

  • Use the original recording, not a re-record. Each conversion loses fidelity.
  • Aim for 16 kHz or higher sample rate. Phone-quality recordings (8 kHz) work but produce more errors.
  • One speaker per file is ideal. Vulgate transcribes overlapping voices but they often run together.
  • Avoid heavy background noise. A noisy café will hurt accuracy more than a quiet but slightly distant mic.
  • Stay under two hours per file when possible. Long files transcribe fine, but splitting them into chapters makes navigation and citation easier later.

Languages

Vulgate auto-detects the language. Latin, Italian, Spanish, French, German, Portuguese, and English are all well-supported, along with most major European and Asian languages. If a speaker switches between languages mid-recording, the transcript will follow them — but you'll get cleaner results by uploading separate files when possible.

What the transcript looks like

Once transcription finishes, you'll find the audio in your Library alongside any other documents. The body of the "document" is the transcript, segmented into paragraphs. The original audio file is preserved and playable from the document viewer.

Transcripts feed into all the same downstream features:

Search help