Bulk transcription

Bulk transcription (Pro) runs one set of settings over many files and exports each transcript as soon as it’s done. Point it at a folder and it can keep watching for new files, which suits voice recorders, podcast archives and lecture folders.

Create a bulk transcription

1. On the home screen, click Bulk Export. The Create Bulk Transcription sheet opens.
2. Choose the source with the Folder / Files switch (see below), then click the button next to it to pick a folder or files. You can also drag files or a folder onto the sheet.
3. Pick the Model and Language.
4. Set the speaker and export options.
5. Click Create Bulk Transcription.

The batch appears in Bulks in the sidebar, with each file as its own job. Jobs run one at a time.

Folder or files

• Folder (the default): TranscribeX finds every supported audio and video file in the folder, including subfolders, and skips hidden files. The sheet lists what it found.
• Files: pick individual files. You can add more later.

Supported formats are the same as for single files: MP3, WAV, M4A, AAC, FLAC, OGG, WMA, AIFF, MP4, MOV, M4V, MKV, AVI, WEBM, WMV and FLV.

Note: The folder can’t be your recordings folder (Settings › General › Save Destination). For recordings, use the meeting automation settings instead.

Watch a folder for new files

With Folder selected, turn on Auto-detect and transcribe new files added to this folder. While TranscribeX is running, it checks the folder every few seconds and adds new media files to the batch, transcribing and exporting them with the same settings.

Files added while TranscribeX is closed are picked up the next time you open it. Keep TranscribeX running (it can sit in the menu bar) if you want new files handled as they arrive.

Options

Speaker Diarisation

Turn on Enable speaker diarisation to label who spoke. It uses all your saved voice profiles automatically. Models that can’t produce timestamps (Qwen3 0.6B and 1.7B, Cohere) skip diarization.

Export Formats

Tick any of Text (on by default), PDF, CSV, Markdown, SRT, VTT and JSON. Each finished file is exported in every format you pick.

• Text Export and PDF Export: choose Text (flowing paragraphs) or Segments (one line per segment).
• Include Timestamps (on), Include Milliseconds, Group by Speakers (on).

Export Location

Export to source location (same folder as media files) is on by default, so each transcript lands next to its audio or video. Turn it off and click Select Export Location to collect all exports in one folder.

Edit bulk settings

1. In Bulks, right-click the batch and choose Edit Bulk Settings.
2. Change the model, language, speaker or export options. For a file batch you can add more files. A watched folder can’t be changed.
3. Click Save Changes.

Existing jobs stay as they are. New settings apply only to jobs added after you save.

If you edit a transcript later, Settings › Automation › Keep auto-exported files updated after edits (on by default) rewrites its exported file, including bulk exports.