Skip to main content
The Transcribe page turns recorded audio and video into text: interviews, podcasts, lectures, voice memos, recorded calls. Each file becomes its own transcription that you can edit, tag, search, play back and export.
The free plan includes 10 file transcriptions in total, one file per run. Pro removes the limit and adds multi-file runs, speaker identification, watch folders, deleting, and export.
The Transcribe page with transcriptions grouped by month, tags on each row and a failed transcription

How to transcribe a file

1

Open the Transcribe dialog

Click Transcribe at the top right of the Transcribe page, or click the banner at the top of the list. You can also drop audio or video files anywhere on the page.
2

Add your files

Drop files into the dialog or click Drop files here or click to browse. With Pro, click Add more files to queue several at once.
3

Check the options

Transcription Model and Language show the current choice. Click Change to pick another for this run only. Turn on Identify Speakers to label who said what (shown only for models that support it). Add Timestamps keeps a clickable time on each segment; it is hidden while Identify Speakers is on, because speaker-labelled transcripts carry their own segments. Notify me when complete shows a banner when the file is done.
4

Click Transcribe

The button reads Transcribe files when you add several. Each file appears in the list as soon as it starts.
The Transcribe dialog with the File and Link tabs, a drop zone, and the Transcription Model, Language, Identify Speakers, Add Timestamps and Notify me when complete rows To transcribe a file from the web instead, use the Link tab. See Transcribe from a link.

Other ways to send a file on macOS

You don’t have to open the dialog first. Each of these opens the Transcribe dialog with the file already added, so you can still pick the model and options:
  • In Finder, right-click an audio or video file and choose Share › Vowen.
  • Right-click a file and choose Open With › Vowen, or drop it on the Vowen icon in the Dock.
  • Drag a recording out of Voice Memos, Photos or Mail onto the Transcribe page.

Queue and processing

Files run in a queue. Local models work through one file at a time. Cloud models run up to four files at once. Waiting rows show a Queued badge and their place in line, for example interview.m4a · #2 in line. Vowen compresses and splits long files for you, so you never need to cut a recording by hand. See File size limits. A file that fails stays in the list marked Transcription failed. Open it, or use Retry transcription on the row, to run it again.

Transcribe settings

The gear next to Transcribe (Transcribe settings) opens three tabs. Transcribe settings on the Defaults tab When auto-export has written files for a transcription, its detail page shows an Open Export button.

The transcription page

Click a row to open it. The page shows an audio player, the title, the date, the speakers and tags, then the transcript. The right sidebar has a Details tab and, for transcripts with speaker labels, a Speakers tab. A transcription with a waveform player, speaker-labelled segments with timestamps, and the Details tab showing Model, Identify Speakers, Add Timestamps and Regenerate Transcript

Rename

The title starts as the file name. Click it, on the detail page or in the list, and type a new one. Changes save as you type.

Play back the audio

When the original audio is available, the player at the top lets you:
  • Play and pause, or press Space.
  • Skip back or forward 10 seconds.
  • Click or drag on the waveform to scrub.
  • Change the playback speed from 0.5× to 2× with the speed button, or with < and >.
  • Click a segment’s timestamp to jump there.
The segment being played is highlighted and scrolled into view.

Edit the transcript

  • Edit text. Click into a segment and type. Undo with Cmd + Z / Ctrl + Z. Redo with Cmd + Shift + Z / Ctrl + Shift + Z or Ctrl + Y.
  • Split a segment. Place the caret where you want the cut, open the segment’s Segment actions menu and choose Split at caret.
  • Merge two lines. Hover the gap between two lines and click Merge, or choose Merge with line above from the segment menu.
  • Delete a segment from the same menu.
  • Move a line to another speaker. On transcripts with speaker labels, use the segment menu to reassign one line to another speaker or a New speaker.

Find and replace

Press Cmd + F / Ctrl + F in the transcript to open the find bar. It highlights every match and shows a count. Enter goes to the next match, Shift + Enter to the previous one, and Esc closes the bar. Click Toggle replace to add a Replace with field. Replace (Enter) changes the current match and All (Shift + Enter) changes every match.

Name the speakers

Open the Speakers tab to rename speakers. The hint reads Rename to update the transcript. Type a name for each speaker, then click Save Changes, or Cancel to discard your edits. To fold one speaker into another, see Speaker identification. The Speakers tab listing each speaker with a name field

Regenerate the transcript

The Details tab holds the options for a new pass over the same audio:
  1. Pick a Model.
  2. Set Identify Speakers, Number of Speakers and Add Timestamps as needed.
  3. Click Regenerate Transcript.
Regenerating replaces the current transcript, including any edits and speaker names. It needs the original file. If the file has moved or been deleted, the tab shows Original file not available.

Ask about the transcript

The box at the bottom of the page (Ask about the transcript…) sends questions about this transcript to your AI model. Use it to pull out quotes, action items or a short summary. These conversations are not saved. See Chat.

More options

The More options menu (…) holds Export, Copy and Delete. Export and Delete are Pro. See Share and export. The More options menu on a transcription with Export, Copy and Delete

Share

Share, at the top right of the page, opens a share sheet with a preview card of the transcript. Drag the card into any app to drop it in as a PDF, or save it as a PDF or Markdown file. Transcriptions can’t be published as public links. See Share and export.

Find and organize transcriptions

  • Search. The search box (Search transcriptions…) matches words anywhere in the transcript text and highlights them. Press / or Cmd + F / Ctrl + F to jump to it.
  • Months. The list is grouped by month, with a count on each group.
  • Tags. Click add tag on a row to add a tag. Transcriptions have their own tag list, separate from meeting notes.
  • Filter. Click tags in the filter bar to show only matching transcriptions. With two or more tags selected, Match: switches between Has any tag and Has all tags.
  • Sort. Choose Newest first, Oldest first or Recently tagged.

Select several transcriptions

Hover a row and tick its checkbox. Shift-click another row to select everything between them. A bar appears at the bottom of the window: Rows that are still processing can’t be selected. Selected transcriptions with the bulk bar at the bottom of the window

Supported formats

For video, Vowen extracts the audio track before transcribing.

Timestamps and subtitles

Most local and cloud models return a start and end time for each segment. Those times power the clickable timestamps, the highlighted segment during playback, and subtitle export (.srt and .vtt). Some models return plain text with no timing. Their transcripts can’t be exported as subtitles; use PDF, plain text or Markdown instead.

File size limits

Vowen compresses and splits large files automatically. What it does depends on the provider.

Troubleshooting

The selected model can’t identify speakers. Pick a model that can, such as Parakeet, Whisper or a cloud model with speaker labels. On the free plan, speaker identification needs Pro.
The transcript has no per-segment timing. Regenerate it with a model that returns timestamps, with Add Timestamps on.
The original file is no longer where Vowen recorded it. Move it back, or transcribe the file again from the Transcribe dialog.