Transcription is simply the process of converting an audio file into a written format so that is can be represented in text, print, script, or video subtitles. There are three types of transcription: verbatim, edited, and intelligent (also known as "clean").
Verbatim Transcription includes any and all imperfections that may exist in the audio, including stutters, sighs, coughs, sneezes, accents, repetitions, and emotional descriptions (i.e. laughing, crying, etc.) This is used in court proceedings, police investigations, interviews for legal purposes, market research, university research programs, psychology, data analysis, and other mediums where direct accuracy is of paramount importance. Full Verbatim includes every detectible decibel with zero omissions. True Verbatim includes all except stutters and pauses that may bog down the reading of the transcription. Intelligent or "Clean" Verbatim includes every sound (including the background noise) minus stuttering and repetition.
Edited Transcriptions are more reader-friendly. These forms of transcription can be used effectively in interviews intended for publication, speeches, conferences, seminars, some business communications, and meeting minutes.
Intelligent Transcription is the file's most polished form. The content is not only edited but also made to be grammatically correct and presented in full sentence form. This is best suited for business communications, interdisciplinary research, the medical field, and some interviews intended for publication.
These three types of translation can be used in isolation or combination depending on the needs of the task(s) demanded by qualitative research, business communication, publication, translation, or other related fields.