Recording high-quality audio for transcription takes a little bit of planning…
Of course, these days you might be able to manage with your smartphone and some solid location selection. But, even then, you can’t just expect to hit record and get sent a perfectly transcribed version later on. Here is how to ensure your transcription services always give you the output you’re looking for:
Learn a little bit about how transcription services work
Before you get started, it’s worth taking a little time to learn how the transcription process works.
Even if you regularly have your podcasts, interviews, market research or webinars transcribed, it can be easy to take the process for granted.
Take a quick glance at our article on the different types of transcription and see for yourself:
Knowing the sorts of factors which will affect your final transcription can help you prepare the way. This will help you always get the kind of transcribed outcome you need for the purpose you have in mind…
Recording high-quality audio – tips and tricks
Problems later in the transcription process are almost always a result of the original recording being of poor quality. Here are six high-quality audio recording tips to ensure yours are always up to scratch:
Choose the right equipment
The audio recording equipment you use will have a huge bearing on the outcome. Pay close attention to:
a) Your microphone
You should always aim to use the highest quality microphone you possibly can. This device more than anything else is going to limit or boost the quality of the audio recording you produce.
Look for a microphone which:
- Is padded
- Has some gauze covering (these eliminate the popping sounds in some speech and also keep the mike hygienic as they can be replaced)
- Has high-quality sound capture
You should consider whether you need a microphone which is:
- Unidirectional: these only capture sound from one direction. They are suited to situations when you are only recording what a single person is saying. Or when you have multiple microphones available, one for each person speaking.
- Omnidirectional: these capture sound from all angles. For situations where you want to record what multiple people are saying, they are often a better choice. If you use an omnidirectional mike to record only one person, you’ll get a lot of unnecessary background noise.
- Switch function: some mikes can be both. They include a switch you can flick between the two. Always be sure to choose the right option for your current situation.
To get the best quality, consider a professional set-up with multiple microphones and a mixing desk where you can control the levels of each.
Some equipment will even allow your transcriber to adjust the levels when playing it back later. This is incredibly useful – especially when transcribing multi-speaker sessions.
b) Your headphones
Listening to the audio through some padded, comfortable headphones while recording will:
- Let you hear the exact quality of audio you are creating
- Stop you from getting distracted by other sounds.
c) Your audio limiter
After first using an audio limiter in audio recordings, you will never want to go back.
Audio limiters work by eliminating sounds outside of a certain range. This means that sudden outbursts (explosions of noise) and over-pitched sounds will be a thing of the past.
d) Your recording software
You don’t necessarily need this. But good software will hugely improve the quality of your recording – especially if you want to edit it, which you almost certainly will.
When selecting recording software, check to make sure yours offers:
- Editing ability
- Noise reduction capability
- Playback capability
- Audio pitch adjustment capability
- Compression capability (without affecting quality)
Create the right environment
Consistency is your goal when it comes to creating an environment for recording.
A studio will be your best bet, of course. There will be no unexpected or unnecessary background noise and the walls will usually be soundproofed or padded, eliminating the echo effect.
But, failing that, anywhere there is no wind (outdoors should always be a last resort) and no excess noise like traffic, loud voices, running equipment (turn off computers, air conditioning and other devices if you are recording in an office-type environment) or background music should be workable.
If you are using a computer as part of your recording equipment, be sure to make sure the microphone is as far away from it as possible.
Your environment is of particular concern if you are going to be using non-ideal equipment, such as your mobile phone.
Finally, of course, you should turn off mobile phones you aren’t using to make your recording or set them to silent.
Hanging a sign on the door to say you’re recording if you’re using a shared space is probably a good idea too!
Place and test your equipment
You will want to place your microphone or microphones as close to the interviewees as possible.
Then you’ll need to make sure everything is working correctly – it’s best to do this before you do the actual recording. Take a test recording and listen to the sound quality. Try moving the mike if you think it could be better, or turning off nearby equipment if you hear a buzzing.
You should also get used to starting and stopping your recording equipment smoothly if you haven’t used it before.
A final step will be to ensure that you know how to get the files from your recording device ready to be sent off to your transcriber without deleting or distorting them.
Lay some ground rules
Next up is to give your participants a few hints to get the best audio recording from them. You can also give them little reminders throughout the session.
Both you and they should attempt to:
- Talk towards the microphone
- Speak slowly and clearly (especially if they have an unusual accent – see below)
- Avoid talking across each other (it’s very difficult to transcribe and you risk missing important details of what was said)
Introduce the speakers
Something which a lot of people overlook when producing audio recordings for transcription for the first time is to identify who is speaking.
You should always start the recording with a round of introductions – get the participants to state their name, where they are from, their job title if applicable and spell out any complicated names.
If interviewees have similar voices or accents, a longer introduction phase can help a transcriber become used to the nuances which make identifying their voice easier.
If any participants in a group setting speak very little, it can be useful to have the interviewer or moderator identify them by name after they have spoken.
A great practice – if you have a spare pair of hands available – is to have someone make notes of speech order. All you need to do is:
- Give each speaker an ID, e.g. “Speaker 1”
- Record a list of number IDs which shows the order in which they speak
- Include the first word or two of what they say by each entry
Offer breaksIf you plan to record for more than an hour, it is a good idea to take breaks. If your participants are starting to look a little uncomfortable or tired, you might even want to take a break sooner.
If this is part of your plan, let your interviewees know first. This way, they can plan their use of their energy and concentration accordingly!
When you return after a break:
- Ensure everyone sits in the same seat
- Don’t start until everyone is in place and has settled down in their seat
- Make sure you turn your recording equipment back on!
How to record high-quality audio using your smartphone
You will always get better results when using proper equipment.
However, these days, smartphone technology has reached a point where it is perfectly possible to create functional audio recordings using little more than your mobile phone.
Almost all of the tips for how to record high-quality audio for transcription listed here still apply. Especially things like the placement of your phone (and the integral microphone), ensuring there is minimal background noise, providing introductions and so on.
But when using a smartphone to record high-quality audio, there are some additional points to bear in mind:
- Only suitable for small groups: because your phone only has a single mike, you will need to make sure that all speakers can sit around it in relatively close proximity. This makes it impractical for groups larger than ten people – and sometimes not ideal even for fewer.
- Choose the right app: most smartphones come with built-in recording software. Make sure yours is fit for task. If not, download one of the many free or paid-for apps which are available.
- Test your recording first: testing your equipment is just as important when you are relying on one device to do everything. Make sure you know how to use the app you’ve chosen.
- Never use a speakerphone: the speakerphone setting is never a good choice for audio recordings. Stick to a properly-designed app and learn how to use it.
- Record phone conversation and chat directly: if you are using an app like Skype, for example, you should aim to record directly rather than to record externally.
After you’ve finished recording
Once you’ve actually made your recording, the job still isn’t quite done. You also need to: Save and back up your recordingSave your recording and then back it up to make sure it doesn’t get lost or damaged.
This is particularly important if you’re using a microcassette recorder. These cassettes are very easy to damage, so backing them up quickly is a smart move.
When it comes to types of files, you’ll find that most Language Services Providers (LSPs) will work with:
As well as a huge range of other formats. For example, at Asian Absolute, we very rarely come across file formats we don’t work with.
Send information to your transcription services
There are a few kinds of information which will make life much easier for your transceiver.
Sending this type of information alongside your recording will enable a much faster turnaround on your project:
a) Provide as many details as you can
When detailing your project to your transcription service provider, be sure to include as many details as you can about your goals and the use your recording will be put to.
If you can send copies of planned questions (or the questions you did ask if you have already completed the recording) along with your audio files, this will be a major advantage too.
b) Send notes relating to speaker accents and languages
Your LSP should assign a transcriber with solid experience working with the particular languages and accents involved in your project.
But even the most experienced transcriber may struggle when a completely unexpected language suddenly appears during a recording.
Sending along notes regarding accents and languages along with your project will be a huge help when your LSP is selecting the right team for your transcription service.
c) Offer a vocabulary document
The best LSPs will ensure that your transcriber has relevant qualifications or experience relating to your project’s subject. At Asian Absolute, for example, we always insist on at least five years of industry experience and/or relevant qualifications to a Masters degree-level or better.
But even highly skilled linguists can stumble over unfamiliar surnames or internal company terminology.
The standard practice for unfamiliar words is to transcribe them phonetically. To avoid this, send along a vocabulary document or glossary of terms including:
- Proper names, including of the participants and places mentioned
- Specialised terminology from the field or industry
- Any acronyms which might appear
- Any unusual words which the participants use – especially if you are unfamiliar with them too! (Have participants explain and spell out the word if it’s very unusual or colloquial.)
When your recording isn’t the quality you hoped forInevitably, there are times when something has gone wrong even though you did everything right.
There may also be occasions when you’ve found a very old recording, has been recorded on old fashioned or substandard equipment or in challenging conditions.
Don’t worry, this isn’t necessarily the end of the world…
Your Language Service Provider may be able to help you. All you need to do is raise the issue nice and early when discussing the project with them.
For instance, at Asian Absolute, we regularly:
- Adjust audio channels to enhance clarity
- Convert files into other formats to reduce distortion
- Use noise reduction software to reduce interference, buzzing and hissing
- Use audio compression tools to smooth volume levels
- Use special equipment to slow down and speed up recordings
Understand what’s involved in the transcription processIt generally takes around 2-3 or even 4 times longer than the length of an audio file to transcribe what is being said on it.
Even the most highly-skilled, professionally trained and experienced transcriber can’t work any faster than that.
This means that splitting the recording between several transcribers to work on can be the order of the day when deadlines are very short.
Doing this requires highly organised project management processes. Otherwise, you risk the individual transcription specialists working on different parts of the file losing overall context and potentially even confusing the names and voices involved.
This is part of the reason why it’s so important to look out for things like ISO:9001 quality management standards when choosing your LSP. Allow time for Quality AssuranceThere really is no substitute for having an experienced editor listen to your audio file in full and compare it against the final transcription as a final check.
Only if your timeframe for delivery is incredibly short and cannot be altered should you consider other QA methods.
As with all things related to transcription, it’s worth taking the time to get it right.Do you need more information about how to record high-quality audio for transcription?
Let’s talk. We transcribe hundreds of thousands of words for our global clients every year.
Comment below or contact us with your questions directly.