Automatic Transcription of the Audio | AI transcription

There is no charge for AI usage through December 31, 2026. For use after that date, see the applicable terms. Available AI functions and technical limits depend on the account.

Automatic transcription changes speech into text with timestamps. Examine the result before publication. Names, technical terms, speaker names, and times can be incorrect.

Availability and costs

Before you start, examine the functions, quotas, and costs in your account. Older statements about Advanced Plus, Business Pro, and a price per minute do not apply to all contracts. See the current plan overview. Contact Podigee Support about an existing or individual contract.

An episode transcript and the generation of other AI content are different actions. Before you encode audio again, examine the audio processing costs.

Create and examine a transcript

1. Open the episode and find the available transcription option. Your interface can show it during publication, under Media file, or in the transcript functions.
2. Read the dialog and examine the selected options. Save a default only if you want to use that selection for subsequent episodes.
3. Start the action and wait for the result. A process that has started is not a completed transcript.
4. Open Transcript, if available. Listen to unclear sections and correct the text, speakers, and times.
5. Save the changes and examine the result again. Use the available download action if you want a copy.

These images show earlier editor and download screens. Current labels can be different.

Published episodes

If your account has a transcription option in Update audio file, examine all processing options before you start. Another transcription can replace your corrections. Keep a copy of edited text first. Do not process the audio again only to correct text if you can edit the existing transcript.

Earlier screens before and after publication:

Feed, player, and blog

After successful processing, JSON and VTT files can be available. The RSS feed can link to existing files through <podcast:transcript>. A started job or text in the editor does not show that both files are available in the feed.

After publication, examine the feed, linked files, and transcript in your player or blog. The display changes with their functions and settings. Transcript links in a public feed can give public access.

Earlier player and blog examples, with and without speakers:

Prepare the recording

• Use a clear recording with little background noise.
• Speak clearly. Do not speak at the same time as another person.
• Examine the text even if the recording quality is good.

A corrected transcript helps people read the content and search the text. It does not automatically give a specified search function, time saving, or correct speaker analysis.

Turn off automatic transcription: check the setting first

The podcast default and the choice for one audio processing task are separate settings. You need permission to edit the podcast or episode.

1. In the classic view, open the podcast and its Audio settings.
2. If the automatic transcription checkbox is available, clear it.
3. Wait for the save confirmation.
4. Reload the page and check whether the checkbox remains clear.

On current plans, the checkbox can appear enabled again. The immediate publication path also enables automatic transcription. Do not treat a cleared checkbox as proof that all transcription is disabled.

For an episode, examine the transcription choice in the advanced options of the publication or audio-update dialog. This choice belongs to that processing task. Closing the dialog without confirmation does not save the task. A choice for one episode is not a reliable account-wide off switch.

Do not start publication or audio processing only to test the checkbox. If you need transcription to remain off, contact support before you process new audio.

An existing transcript remains separate

Changing a default does not delete an existing transcript or recall a file already distributed. Keep a copy of edited text before you change the transcript itself.

The checkbox turns itself back on

Send support the podcast and episode links, your plan, and whether the control is in podcast settings or an episode dialog. Include the browser, interface, exact steps, and screenshots before and after saving. State whether it returns immediately, after saving, or after reloading. This needs a setting and workflow check; repeated audio processing is not a fix.