Audio Recordings for Transcription - How To Get The Best Results

Whether you are recording dictation, interviews, conferences, focus groups or meetings, it is important to know how to get the best quality recording, how to find a competent transcriptionist, and what sort of costs you are looking at.

The type of transcription depends on the type of recording equipment you have. You can record on to standard audiotape, micro or mini tapes, mini-disks and digital files, which the transcriptionist plays back on a computer. If you don’t already have recording equipment, bear in mind that while a mini tape Dictaphone, for instance, might be much cheaper then a fully functional digital conference recorder with several external microphones, the quality of the cheaper instrument makes it suitable only for basic dictation.

Some mini and micro tape recorders are powerful enough to be placed on a table between an interviewer and interviewee and pick up both voices, but we strongly recommend you carry out a test recording first to make sure. To record a focus group or conference you will need a good system of powerful microphones which will probably record into a video, digital audio or standard tape system. Digital systems are very variable and many are of good enough quality to record focus groups and conferences.

It is also worth bearing in mind that more and more transcribers no longer offer tape transcription, as digital is becoming so much more popular, and it is easier to work with and better quality.

On average it takes four hours to transcribe one hour of recording, but this varies considerably. It might be closer to three hours for one person speaking into a Dictaphone but could well be at least six to eight hours for a large focus-group.

This is because we speak much faster than even the fastest typist can type. Also, the transcriptionist doesn’t just type what he or she hears - it has to make sense. Although your transcriptionist should only type what’s being said (unless specifically requested to edit), it's not just a matter of the right words, but also the right punctuation and, of course, picking the correct homonym (e.g. there or their; two, to or too; pare, pear or pair; fair or fare; the list is almost endless).

You will also need to decide whether you want a ‘verbatim’ transcription with every um, er etc. transcribed and all interruptions etc. noted. This is necessary for some work, such as police transcription, but is completely pointless in, for example, conference transcription, as the important factors in the latter are readability as well as accuracy.

The time taken to transcribe a recording can vary according to a number of factors. These include: • Clarity of speech - Strong or varied accents slow down the transcription. • Clarity of recording - background hiss will make the speech more difficult to hear and therefore slow down the transcribing. Also, if a number of people are being recorded, the positioning and number of microphones can make an enormous difference to the quality of recording. • Background noise - a recording in a pub or café is likely to have significant background noise and deciphering the speech from the background noise can be difficult. • Number of people talking - everyone talks at a different rate and pitch, and with a different accent. Having to 'tune in' to a number of different people slows down the speed of transcription. • People also talk over each other and interrupt, and in order for the transcriptionist to transcribe two people talking over each other they will have to play back the section of recording several times to pick out the different voices. • Speed of speech - as already mentioned, some people speak faster than others. Clearly a fast speaker will take longer to transcribe, as they will fit more words in to an hour than a slower speaker. • Whether speaker speaks in coherent sentences - few people actually do speak in the same way as we write. There are usually multiple 'ums and ers' and people often speak in partial sentences. This can require the transcriber re-listening to a section several times to grasp the meaning and insert the relevant punctuation. • Whether you are seeking an exact transcription or a transcription 'for meaning'. Our standard offering at Penguin Office Services is to transcribe word for word but to leave out 'ums and ers' and repeated meaningless phrases, such as 'you know', and 'know what I mean?' It is obviously quicker to leave out these parts, but we are happy to transcribe them if required. An example of when an exact transcription might be required is academic research interviews studying language use.

So if you want to make sure you get a good quality recording, what can you do? Firstly, use the best quality recording equipment you can afford. This really is a question of getting what you pay for. A low-cost recorder may seem economical but the quality will be comparatively poor and a poor recording will take longer to transcribe. Paying for more transcription time will probably work out more expensive than spending a few pounds extra on better recording equipment. Digital recording must be the way to go if you are starting from scratch. Minidisks make very good quality recordings but only a few transcriptionists can transcribe from them. At Penguin Office Services we can convert Sony Minidisk recording to computer-based audio and transcribe from that. If you are not going down the digital route then standard audio tapes make better quality recordings than mini tapes, though micro tapes are also good for interviews and one person dictations. If you are using tapes, by all means keep a back-up copy but do send your transcriptionist the originals, as back-ups may be reduced quality.

Secondly, record in a quiet environment such as a self-contained office. This will make an enormous difference to the quality of recording. If you need to record in an open-plan office or public space then try to ensure your microphone is placed close to the speakers you are recording, but as far as possible from other workers in the office and their phones, or other people in a public space. Recording in pubs, restaurants and cafes, trains etc. will greatly reduce the quality of recording. If the speaker has a very quiet voice the recorder will not pick it up. Although you don't want to remind people that the recording equipment is there, you may have to ask them to speak up or place the equipment closer to them.

Thirdly, bear in mind that most built-in microphones are of poor quality with limited control over volume levels. They are designed to pick up close speech such as someone speaking directly into a Dictaphone. Using a good quality microphone for interviews and other multiple speaker recordings will greatly improve quality and reduce transcription time.

It is essential to ensure that the speaker is close enough to the microphone. If there are several speakers, for example in a seminar or conference, ensure that there are sufficient microphones around the room to pick up ALL the participants clearly. If seated around a table use something like a Conference mixer, which links several microphones to one central unit that will pick up a discussion more clearly than one central microphone.

If you are chairing a focus group you should ensure that all participants know that they must speak clearly and one at a time or their words will not be clearly recorded. There will be times when participants get exited and interrupt each other. When this happens it may be sensible to interrupt and ask them to repeat, one at a time, so that all the words will be clearly recorded.

There are also a couple of things you really should avoid doing, and these usually apply to tapes, rather then digital. Some recorders can tape at slower speeds, extending the recording time and reducing the number of tapes used. However, there is a related loss in recording quality. We would recommend that you use only the fastest speed setting on your recording equipment. Also, some recorders have a voice-activation system so that you don't waste tape recording long pauses. However there are two serious problems when using this. The first is that a soft-voiced person might not activate the recorder at all, so that whole sentences might not be recorded. The second is that every time somebody does speak and activate the recorder there is a slight time delay before recording starts, so the first word or two will probably not be recorded.

Finally, where you have speakers at a podium or multiple speakers in a large room, it is advised that you seek professional help to ensure everything is clearly recorded. Most hotels and conference centers will have in-house experts, or a company specializing in audiovisual equipment should be able to help.

If you have decided to use digital recording, there are quite a number of things to bear in mind when choosing your recorder. Please see my separate article, ‘Digital File Types for Transcription’ for more information on this.