Audio and Voice Recording for Digital Storytelling by Simon Turner
When planning your set-up, remember that the way you capture sound demands as much attention as the other elements of Digital Storytelling. The sound, in particular the voice, is the foundation of every film, and will dictate the pace, feel and structure of all visual and editing elements that are subsequently involved.
It is vital, therefore, to get the sound right at the outset with the minimum of fuss and at a good enough quality to work in a variety of mediums - internet, radio, television, CD ROM, etc.
What follows are some basic guidelines to help you achieve this, but because of the varying nature of different software systems and computer configurations that individual teams will use, you will not find any tutorials on specific systems. However, just about all of them use a similar set of protocols, and this section is based on that presumption.
Each team should appoint someone to oversee the audio software and accompanying kit, to familiarise themselves with that particular system, how it works and what works best. They can then disseminate this condensed knowledge amongst the other team members.
When choosing equipment and software, everyone will have differing views on what is best. A reasonable starting point is to ensure that you have a decent microphone; audio interface with a microphone pre-amp and USB or Firewire interface; audio editing software; and a computer. A portable hard drive to transfer data between the sound computer and the editing computers is extremely useful too.
The microphone should be flexible enough to cope with recording other audio such as musical instruments and sound effects, as well as being good for voice work. Good does not, however, mean expensive - the one mentioned later, the Rhodes NT3, cost around £90.00 inc. VAT and is a very good microphone. There are plenty more similar mics out there at competitive prices, so it pays to have a look around. Avoid very sensitive microphones as these will pick up absolutely everything and should only be used in sound-proofed areas. You don't need a microphone that has its own separate power supply. A cardioid polarity (polarity is the way in which a microphone 'hears' sound) is suitable for most eventualities, and getting some shielded cable for it is not a bad idea either.
As far as the audio interface and software is concerned, there are lots to choose from. Most of the interfaces come with their own audio software, so you will need to check these out to find which one suits your needs and capabilities the best. Also, if you want to use a particular piece of software it pays to check that it works with the interface box you use, and vice-versa. They are not always compatible, even if they say they are. Try to make sure, if you can, that the software allows you to edit more than one track at a time as this can be very useful, and do check that it will allow you to record/export your sound files at the correct bit/sample rates (16 bit 48k is broadcast standard).
You can download most of these software systems as demos so do try before you buy.
When it comes to computer hardware, people tend to fall into two main areas - PC or Macintosh - with very little common ground between them. The truth of the matter is that both systems are perfectly useable and it is not an impossible learning curve to move from one system to another if you need to.
It is worth noting that, on the whole, Macs do tend to be more stable when it comes to media production and are geared up to work 'straight out of the box' with the minimum of fuss.
Whatever system you decide on, make sure that there are no compatibility issues between hardware and software, and get a reasonable amount of RAM put into the machine (2Gb or more is ideal).
Above all, make sure you test-drive the system thoroughly before taking it 'on the road'. Any glitches or issues should be dealt with and solved before any workshops are considered, and whoever is covering the sound should be familiar with every aspect of the equipment and system. More often than not you may get warning messages that seem far worse than they actually are, or you may be asked to configure a setting that is unfamiliar but crucial. All these things should have been gone through and dealt with, and a note made so that if it happens again you will know what to do.
Try to make the computer crash if you can, and find out why and how to solve it. All computers crash without exception, and the audio specialist should have a comprehensive enough knowledge of the hardware and software to know why it happens, how to re-boot it, and wherever possible, how to prevent it from doing it again.
The Recording Sequence
1. The Story Circle
Listening to a story is always better when the narrator is telling it from experience. There is a vast difference between someone simply reading something out, and actually telling it with all the attached personal and emotional aspects from their own perspective. This is true for all types of stories, whether comical, sad, strange or droll. This is why Digital Stories can have such an impact on the audience, because they are real stories told by real people.
It is the true voice of the storyteller, not an actor, exposing the genuine feelings and personal reactions involved that gives us a sense of inclusion and emotive insight into a unique event. Getting this from people who, on the whole, are very nervous can be an interesting challenge. Indeed, the first words that someone usually utters when coming to record their voiceover is "I've been dreading this". Why should this be the case? More often than not it is because they assume that they are expected to 'perform', to be able to offer us their tale in the form of a 'professional' voice-over, or they feel that their voice is simply not good enough when compared to those they have heard on radio and television. Of course this is not true, so we must ensure that they understand that their real voice is what we are after, and to be calm enough to ensure that this comes across. So, what can we do to get people to relax as much as possible, to focus and reveal their true voice, and to get the narration done within a reasonable timescale?
The first time that people will come into contact with the notion of actually recording their voice is during the Story Circle, where they are introduced to the additional members of BBC staff and the key tasks they will be performing. It's during this introduction that the audio technician can make an important announcement - "we do not expect any of you to suddenly adopt a Radio 4 vocal style. We are looking to record your story narrated in your own voice". They should also point out that the narration doesn't have to be done in one continuous take, and that storytellers should lose the notion of this being treated as a 'live' event. Of course, doing it in one take can be very useful, but it is not a prerequisite under any circumstances and should not be seen or indeed promoted as a challenge.
Explain what happens during the recording process, how you can keep one bit and re-do another (dropping-in) and how you can insert or close up gaps and cut out errors. Try to answer any questions that may be causing them anxiety but be mindful that you can explain things a lot more easily when they come to do their voice-overs. Above all, put everyone at their ease and make sure that they are aware that it is a reasonably easy process. Its surprising how many people will have been quietly worrying about these factors, and this will help to put them at their ease in the build-up to them actually recording their stories in the audio area.
2. Music and Sound Effects
Before dealing with the voice, it's important to raise the issue of music and sound effects. During the Story Circle, after each story is read out, music and sound can be brought into the equation and this is the point when you can find out what they want for their film, and any suggestions you have can be brought up, discussed within the group, and noted for later.
It is useful to do this in the Story Circle, as other participants will be inspired to think more about sound and whether or not they need any for their films, and will be made aware of the potential advantages and problems associated with this aspect of their films.
It's always best to encourage them to use music that either they or a friend have created: even the simplest thing can be effective - whistling, humming, singing a pertinent song i.e. a school hymn. Sometimes they will have brought in CDs or tapes of favourite music they want to use. You will have to check copyright on every piece of pre-recorded music that you use without exception. If you do not check, you will be legally obliged to pay a fee and in some cases may even be prosecuted. Also, just because it is a favourite piece of music does not mean it is right. Music that reminds one person of a happy/sad/important time in their life can mean something very different to others, and this can alter a story's meaning, sometimes very detrimentally. Also, does it match the story's context, and will it be interruptive when placed under the narration?
Another good technique is to see if they have any connected audio material that they have brought. One such story was made by a reporter who was present at the scene of a bombing. At that time, she had made recordings of the immediate aftermath. We underscored her narration two thirds of the way through with the sounds of people caught in an awful moment, simply increasing to a crescendo and ending the film with a man screaming, "These are my people". Finding effective punctuation points in music and sound is as important as getting a good voice-over, and can add great depth to a story as long as it is never given greater emphasis than the narration.
More often than not music and sound is wanted, but not thought about. We can discuss, suggest and facilitate where necessary, and if the team is in a position to do so, even create original music if required. Bear in mind, however, that music/sound should only be used if it serves a useful purpose. If in any doubt as to whether it works or not, or is needed, always leave it out. The voice is the most important element of audio, and anything that dilutes its effectiveness or interrupts it will only serve to weaken the story.
The important things to remember with this area are:
The soundtrack/sound FX should not fight with the voice
Well-known pre-recorded music will have strong connotations that will very likely distract from the story and add unintentional meanings
Its always better to get them to provide music created either by themselves or a friend - whistling, humming, singing, playing an instrument (within their level of ability - very often a couple of chords can be just as meaningful as a well played, complicated piece) are all very potent in the context of the stories.
If they want to record something at home or elsewhere to use in their film, discuss the best way of achieving a good result. Explain about levels, mic positioning etc. Also, if they are using your kit, be aware that accidents happen, and ultimately it is your responsibility if you lend out equipment to participants without supervision.
Where possible, go with them to help this process, or get them to bring the instrument/sound source/musician into the workshop and record it there. Please be aware that going off-site should always follow health and safety criteria, and there should always be at least three people in the group. The workshop leader must be made aware when participants and staff go off-site, and must be informed as to where they are going, what they intend to do, and how to contact them if required.
Always discuss everything you are doing with the participant. This is their story, and they must be made to feel that they have involvement with the entire process.
Any changes made or problems raised concerning their soundtrack, even subtle ones, should always be facilitated either with them, or explained clearly to them. This includes things like volume changes etc.
Sometimes the inclusion of music and sounds can dilute the story - it is as much about when to leave it out, as it is to add it.
The simpler the additional audio, the stronger the voice will remain.
This is not about personal taste - does their suggestion not work in your mind because you don't like that particular piece of music or sound?
What audio artefacts or simple skills (i.e. whistling) do they have already that could be used to embellish their story?
3. The Voice
Narration traditionally falls into two distinct areas: informative, which offers extra dimensions to what we are seeing, and explanatory, which contextualizes what we are seeing on screen and why.
The extra levels to narrative include the emotional tones of the persons' voice. Does their story sadden them? Are they amazed by it? Do they find it funny? Whatever the underlying feeling, it is this that we are trying to bring out in the most honest way.
There are some technical points to cover before going any further. The area designated for recording can range from a really good radio studio to a toilet, and as such you have to be ready to accommodate all assorts of acoustic situations. Not only that, but of course the voices themselves differ wildly in timbre, volume and sibilance.
To get a good uniform feel to all the voices both in level and overall tone, you can use an external unit to compress, limit and de-ess (remove sibilance from) the voice before it reaches the computer. This not only helps with an overall level balance at source, it also allows us to "tune" the mic to the room, add a little presence if needed, and generally tweak the sound till we are happy with it. These facilities are also usually available as part of the audio software packages - the rule of thumb if using either is to not overdo it, and to apply these processes subtly. If you don't have these facilities, then a pop shield can often help variations in levels. Indeed, you should always use one anyway to get rid of plosives - consonants that cause the 'eardrum' of the microphone to overload and cause a 'popping' effect. A good rule to remember is to get the microphone reasonably close to the sound source, and to record at a strong enough level. This will ensure that you will record more of the original sound (in this case the voice) and be able to play it back without having to turn it up, consequently increasing the level of background noise.
You will only really get to know your kit and what works best by trying it out. The microphone we use is a Rhodes NT3. This has a cardioid polarity (meaning it has a heart shaped response, or way of hearing), and is a condenser microphone (which means it has to receive power from either a battery or phantom powering from the audio interface, mixer etc.) It is inexpensive and good for pretty much everything, but there are many other similar products to choose from that do an equally good job. Just remember that you don't have to spend a fortune to get a good microphone, but you can spend a lot on something that is not suitable and will cause more problems than it solves. In-built camera microphones should not be used, as they rarely give good results, and radio or clip mics (if available) are usually used only in circumstances such as outside location recording etc. but can be used if that's all there is.
Practice and experiment and make notes on what works best before taking it out on the road/doing your first workshop etc. Also, be aware that you will never achieve absolute uniformity between voices, so just get the best sound you can from each. Please remember that a badly made recording is very difficult to fix. There is no magic formula for fixing poor sound, so get the best recording you can from the start.
4. The Recording
A key issue is the script and certain words or phrases. What can look good on paper in the main room can sometimes sound effected and false when read out in the audio area. If this is the case, ask them how they would phrase that sentence what word would they use in normal conversation. This will lead to a more naturalistic sounding narration and will be easier to read out loud.
When the participant comes in they will no doubt, as highlighted earlier, be nervous. From the moment they come in, it's important to make them feel comfortable. Tell them (and ensure) that there will be two of you with them in the room. As per recording on location, it is vital that two members of the team are present in the sound room, and providing both team members are sensitive to the situation, this should not cause a problem.
Tell them what your role in the process is, and what you are actually doing. This makes you part of the voiceover, and takes the focus away from them and turns it into something you will be doing together.
Its always polite to ask them if they would prefer to sit or stand at the start, and if after a while things aren't working, get them to do the opposite. Also check their body language. If they are gripping the seat, have their arms folded tightly across their chest, aren't breathing normally etc, point it out. Relax them, do some breathing exercises, get them to walk around the room, get them some water, tell a joke, whatever it takes. Don't get impatient, but don't be afraid to push them if it's needed.
Do the first take as a "checker". Tell them that you will record it but it is purely for getting the levels and settings right. This is only partially true, as it also helps "break the ice" of telling their story out loud in the sound room.
The second take should be done without any advice from you. Just tell them to relax, and to tell the story in their own time. They will be more than likely aware of the time limit on their film, and will be worrying that they are going to go over this. Tell them that it doesn't matter - the gaps can be closed up later - and that if they fluff a line to just go back to the start of the previous sentence and, in their own time, start again.
Remind them what is needed and don't ever be negative. Suggest improvements in a positive way, and discuss with them the meaning and emotional importance of what they are saying and how best to get it across. When you feel you are getting somewhere, it's often a good idea to play them the take and to discuss the areas you feel could be even better. Having an example is far better than just telling them, and although you shouldn't do it for every take, once in a while is useful. Remember that most people dislike the sound of their own voice when it is played back to them. Be supportive and encouraging as well as answering any concerns that they may have.
Record every take. Only delete the takes you don't want when you are sure you have what you need.
Ask them how they think it's going - comment honestly on their responses and talk to them about their experience so far. This can help them relax and feel less pressured. The second team member should feel able to add input, but remember that too much instruction all at once from two people can be overwhelming. These things can take a little while so one voice at a time and any advice given should have a reason given to accompany it.
Always stop if they are getting tired. It's far better to come back to it later than to repeat the same flat narration. Judge when they have had enough and judge when you think you have got the best you can from them. Some people just don't have very animated voices, and no technique in the world is going to prise it from them.
As far as technical requirements are concerned, the usual broadcast standard applies if this is going onto television - 16 bit, 48k. However, the usual -6dB rating can be ignored, as most video editing software including Premiere is a bit fiddly to work with if you have lower signal levels.
It's not a bad idea to keep to these standards as a general working practice, and convert the audio for other uses such as the internet etc. afterwards.
Once the voice-overs are all captured, it is important to check each one over loud speakers. You will hear any glitches or defects and can fix them there and then. You can also do a level comparison, and normalize and then alter the gain of each voiceover to get them all sounding reasonably similar. Normalising is making sure the loudest part of the soundtrack does not exceed a preset limit - again, usually part of the audio software package.
Remember that although everyone's voice is different; you are trying to get each voice level sounding as natural as possible. If someone's voice is quiet, then trying to make it loud will sound unnatural. A little bit of volume will help, and providing you recorded it at a good level at the start you should have no problems.
Normalizing the voiceover to -1dB seems to work very well, and provides a good visual waveform in the time line. The levels, providing they remain no higher than -1dB, can be uniformly lowered/raised as needs be in post-production after the workshop.
As far as EQ is concerned, once the voice is recorded it should be left alone. The tonal quality of the voice should have been captured at source the way you want it to sound in the film. Mucking about with frequencies after the event, unless you know what you are doing and why you are doing it, will just cause time-consuming problems.
Audio effects, such as reverb, delay etc, can be great fun and can sometimes enhance a soundtrack, but if you use a 'long' reverb for example (one that has a long decay), then this can sound wrong. Any effect should be used sparingly and, just like music, if in doubt leave it out.
If there are unavoidable loud background sounds on the voiceover, and you need to make cuts between sections of dialogue, a good idea is to put a volume fade at either end of each section. In other words, rather than cutting the sound dead between each section, allow the voice to finish speaking, and then a quick and subtle volume fade-out to silence can help considerably. The opposite applies to the beginning of a section - a quick fade up from silence before the voice starts speaking can smooth out most problems of traffic, distant crowd noises etc. If you have to do this, try and make any edits or changes to the soundtrack as subtle as possible. Bear in mind also, that any small glitches such as clicks, hisses and so on that sound quiet to you in the audio room, will be increased to very noticeable levels in the final showing of the films. These issues must be solved before transferring the voiceovers onto the participants' machines. The same applies for any music or other audio that is to be used.
6. Final Points
One thing to mention is that any extras such as music should always be put on to the participants' machines as a separate sound file. If you do the voice and music/SFX as a finished mix this does not allow for individual editing of the various audio elements, and again can be the cause of problems.
Once the voiceovers and music are in the participant's machines, it's important that you keep coming back from time to time, and checking what they have done. Again, if they have created an odd edit or it sounds wrong, discuss, suggest and include them in the process of amendment. Explain why you think it would be better a different way.
Always check the films before they are dubbed onto tape for the final showing. Some people can't help "tweaking" their finished films.
Finally, remember that once these films are made, they cannot be 'un-made'. Both the experience of the workshop as well as the final film is what will matter most to the participant, and the voiceover should be something that they feel proud of as well as conveying the correct meaning to the audience.