How to transcribe text and audio

The term “transcription” is used to describe the act of transcribing an audio file recorded during an event or conversation.

The Italian word for transcription (sbobinatura) derives from the reels of magnetic tape used to record audio before it was transcribed. This term has remained in use in Italy even though technological developments have rendered tape recorders and reels of magnetic tape obsolete. It’s a bit like what's happened with floppy disks, which become the universal icon for “saving” digital files.

What does the transcription of audio files involve?

As previously mentioned, audio transcription involves transcribing spoken words on an audio file, and there are various levels and types of transcription depending on the need at hand.

Basic transcription is limited to the word-for-word transcription of a text, omitting all punctuation except for full stops. The  grammar is not corrected, resulting in a transcription that perfectly reflects what the speaker(s) has said.

The act of refining a transcript is referred to as intelligent transcription. This service involves correcting grammar, adding punctuation to make the text easier to read, identifying speakers, and adding time stamps.

Verbatim transcriptions aim to enrich the text with additional information that helps the reader to better understand what happened during the recording. This service includes the addition of ambient noise, moods and tone, overlapping voices, and so on.

The difficulties that come with audio transcription

The ultimate goal of transcription (be it advanced or any other kind) is to create easy-to-read texts that can be saved, used and distributed. For this reason, particular attention must be paid to the quality of the transcript produced, especially if it is destined for public use.

The main difficulty that comes with producing a transcription involves overcoming audio quality problems that can at times render certain parts of speech unintelligible. When transcribing audio in a foreign language, problems relating to accents or the use of technical or specialist vocabulary can present difficulties, especially if the transcriber would have similar difficulties transcribing the same audio file in their native tongue.

Regardless of the language spoken, a well-executed transcription results in an accurate transcript that is well-formatted and easy to read.