A transcript is a text version of dialogue and/or non-spoken audio information from a video audio recording. Descriptive transcripts also include text description of the visual information needed to understand the content. Transcripts are a good idea for audio-only content like podcasts, Twitter Spaces, and Clubhouse rooms.
According to the speech-to-text service Rev, there are three different types of transcripts: full verbatim, clean verbatim, and detailed notes.
A full verbatim transcript doesn’t remove any false starts or verbal tics like “ums” or “likes”. It transcribes every word and utterance as they occurred in the original media selection.
A clean verbatim transcription removes the verbal tics and false starts and is also known as an “intelligent verbatim,” or “word for word” transcript. It does not alter the order of dialogue, just cleans it up. Most people probably don’t want to read every misspoken word or “um,” so a clean verbatim transcript is normally preferred.
Detailed notes distill an audio or video recording down to its essential parts without forcing the viewer to read large chunks of text.
Rev has a great synopsis of how to create well-formatted transcripts, which is summarized below.