How To Accurately Transcribe Audio And video recordings for court use

Audio and video recordings are often submitted as evidence for court proceedings. In Singapore, under the State Courts Practice Directions 2021, Part XIX, Paragraph 130, regarding the documentation of audio and video recordings for the community courts and tribunals cluster, “A transcript of the important part(s) of the audio or video recording must be prepared to state the relevant fact (ie, the action that is seen, the exact words that are used, the type(s) of sound(s) that are heard).”

Although this does not have to be done by a transcription service provider, legal firms often refer the task to one to save on time and to ensure quality.

If you would like to know how an accurate transcript is produced, read on.   

Accuracy

For audio recordings, repeated listening is key to producing an error-free transcript. This is to ensure words are heard correctly, no word is missed out, and overlapped speech can be differentiated as best as possible. We recommend listening to a recording at least three times, especially for difficult ones, such as those containing muffled words, overlapped speech, or an unfamiliar accent. It can also help to have more than one person review the recording.

Maintaining objectivity is also very important. Any subjectivity or interpretative quality to a transcript may be deemed biased or unfair in court, rendering the evidence useless. For example, if a person in a video is seen smiling, you should describe the person as “smiling”, but not “happy”, as the person who is smiling may not necessarily be happy. Hence, it is important to maintain objectivity and to avoid introducing subjective interpretations when transcribing.

Formatting

Apart from being accurate, a transcript must also be easy to follow. A range of formatting tricks will help you to do this. Let’s discuss some of them.  

Before you start, know what style of transcription you would be using--full verbatim or clean verbatim. Full verbatim retains filler words and stammering while clean verbatim presents the speech in Standard English.

You may also want to consider other formatting elements that will help you present your transcripts clearly or that may be relevant to meet your needs. Some suggestions are listed below.

·         Headings and paragraphs – Sectioning the document into different parts with headers and splitting up large paragraphs into smaller ones can help improve readability. 

·         Speaker labels – Speaker labels are required to differentiate between different speakers. If a speaker’s identity is not known, generic labels like “Male”, “Female”, or “Speaker 1” can be adopted. 

·         Timestamps – Placing timestamps helps the reader to locate parts of the recording that correspond to particular places in the transcript. Timestamps can be inserted at regular time intervals, for example, every 5 minutes, by inserting [0:05:00], [0:10:00], etc. at the relevant points, or may be inserted every time a different speaker begins their turn. 

·        Tags for inaudible speech and crosstalk – Tags like [inaudible] and [crosstalk] can be used to indicate instances of inaudible speech or crosstalk. 

·         Describing non-speech sounds – Non-speech sounds can be indicated by enclosing them within brackets. For e.g., [clapping], [door closing], etc. 

·         Describing actions – For video recordings, you may italicize or enclose actions in brackets. For e.g., [Mike walks over to the table].

There it is! With these tips, you are on your way to producing a solid transcript.

Alternatively, you may leave the job to a professional company such as ours.

Parts of the recording that require translation

What if your recording contains parts that are not in English? According to the rules stated in State Courts Practice Directions 2021, when words in the audio or video recording are not in English, a certified translation of the words into English must be provided.