Recording Interviews for Transcription

Audio quality is once again the topic today. When one considers the importance of the input gained from each research participant, some discernment must be applied to ensure that the recording is useable in terms of its viability for transcription. Researchers are often at the mercy of their participants when arranging interviews, and these sessions can therefore take place in environments that pose significant challenges to producing a recording that is optimised for efficient and accurate transcription. A staff canteen, an airport lounge or a factory floor are certainly not optimal environments in which to conduct a research interview that is to be transcribed. (We have in fact received recordings of interviews that took place in a factory.)

An audio file recently received presented a very disappointing situation to the researcher. The interview was conducted in a French-speaking country and therefore the participant spoke English with a French accent. In this particular audio file, the participant's English could be described as "broken" but probably more accurately described as "shattered". If this interview had taken place in a quiet office it would have been "doable" but this definitely was not the case. The interview took place in a truck. From what could be discerned, not only was the English very broken, indeed so was the truck. To further exacerbate matters, the truck was in motion along a dirt road for the duration of the 75 minute trip.

To sum up, we have a recording containing an interviewee speaking broken English with a very broad French accent, travelling in a broken truck along a dirt road and ... the radio is turned on, volume turned up and a chat show now contributes to the already ridiculously noisy cabin environment. Of course, a helpful and comprehensive transcript cannot be produced from audio of this nature.

Researchers need to give cognisance to the fact that the information imparted by the participant is lost if due consideration is not given to optimising the recording for transcription.