Skip to main content
Add-ons are made from the transcript. Switch them on with config.addons when you create a transcript, or add them later to a completed transcript without transcribing it again.

Which add-ons are there and what do they cost?

Prices are per minute of audio, on top of the model price. Translation is charged per language.

How do I switch add-ons on?

Send config with POST /transcripts:
translation needs translate_to, 1 to 5 language codes. custom_fields needs config.custom_fields, 1 to 20 questions, each with a key, a question and a type (text, number, boolean, list, speaker or choice).

How do I add add-ons to a finished transcript?

Call POST /transcripts/{id}/addons with addons and, if needed, translate_to or custom_fields. The difference in per-minute price is charged for the recording’s length.

When are the results ready?

Most add-ons are generated shortly after the transcript. Each has a status in the result: pending while it is being made, ready when done, and failed if it could not be generated, in which case it is refunded. The transcript.addons.completed webhook fires when they are ready. Subtitles and key terms are available with the transcript.

What does redact_pii do to the transcript?

With redact_pii the transcript stays processing and the result returns 409 until personal details have been looked for, so nothing unredacted is served. Redacted text uses typed tags such as [PERSON_NAME] and [EMAIL]. The result lists the number of redactions, their types, and audio ranges you can bleep.

How do speaker names appear in generated text?

Generated text refers to speakers as {{speaker_1}}. Replace each placeholder with the name in speaker_names before you show it.
Last modified on October 26, 2026