In today’s article, we will focus on inserting timestamps in transcripts.
· Entering timestamps is largely a manual task performed by transcribers who listen to and transcribe the audio files. This transcript is subsequently reviewed by editors who also need to ensure that the final product is of good quality with careful compliance of instructions.
o Being a manual task, it consumes the transcribers’ time and efforts depending upon the frequency of the required timestamps.
· Inserting timestamps at regular intervals takes the transcribers’ attention away from their basic transcription requirements.
o In order to ensure quality output with timestamps and compliance of instructions, the transcribers spend an unusual amount of time transcribing audio files with such instructions. This ultimately results in loss of productivity for the transcribers.
· The following are some of the examples of the frequency of timestamps, i.e. when to enter a timestamp:
o Every 5 minutes, every 2 minutes, every 1 minute, every sentence (suitable for subtitling), every time there is a change of speaker, etcetera. These are just a snapshot of some requests we have received so far.
· The most time-consuming and exhaustive of the above methods is when one needs to insert timestamps in every sentence. Again, if both the start and end time are required, one cannot expect a transcriber to transcribe more than 20 minutes in a particular day (ideally a transcriber can transcribe more than 50 minutes in an 8-hour shift). This indicates that transcribers need to insert a timestamp at the beginning and the conclusion of a sentence, then again at the beginning and end of sentence2, and so on. This process is repeated until the end of the audio file.
o For example:
§ 0:00:05 – 0:00:20
§ Sentence1
§ 0:00:21 – 0:00:39
§ Sentence2
· Consequently, the lesser the frequency, the more time and efforts required.
· Again, inserting timestamps for every speaker change can be complicated.This is because several times, speakers (mostly discussion participants and in some cases interviewers/interviewees) interrupt each other and thus cut off the flow of the conversation as they tend to speak at the same time. In this case, the process becomes even more difficult.
· These days, there are some software products available which claim to help insert timestamps in a transcript. After exploring those products, we have realized that inserting a timestamp is still a combination of two commands. This means that every time a transcriber wants to enter a timestamp, they need to provide two keystrokes. This instance illustrates that manual efforts have to be taken to maintain the quality of the transcript.
· All in all, inserting timestamps at regular intervals, for example, at a frequency of less than 5 minutes, requires meticulous and intensive efforts on the part of the transcribers.