Auto transcribe

Auto transcribe is our most popular option for creating captions. When selected, CaptionHub sends your audio to a speech recognition provider, which returns a raw transcript. CaptionHub then uses its proprietary Natural Captions technology to process the transcript into aligned, human-readable captions.

CaptionHub currently supports five speech recognition providers:

Each provider offers slightly different functionality, all of which is supported within CaptionHub. To change your default transcription provider, open the Default providers tab of Team settings.

To use Auto transcribe, select it from either the Create captions dialogue box or the Replace original captions dialogue box. You’ll then need to choose the language spoken in your video.

You can view the languages supported by each transcription engine here.

Depending on your subscription, you may have access to custom dictionaries. Custom dictionaries help bias speech recognition towards predefined terminology such as names, places, products, or brand-specific language.

AI transcript fixes is an optional step that tidies up your transcript before it is split into captions. Using your team's AI provider, CaptionHub removes full stops placed in the middle of a sentence, turns a full stop at the end of a question into a question mark, and drops the false starts, stutters, filler words and slips of the tongue of natural speech ("we're, we're doing", or "you know" as an aside). Cleaner sentences give you better captions and more accurate translations, because translation errors often start with a badly punctuated transcript.

AI transcript fixes never adds or changes words. It can only remove words and move punctuation, and if a fix would change too much of a passage, CaptionHub keeps that passage exactly as transcribed. It works for English, German, Spanish, French, Italian and Portuguese. To use AI transcript fixes: