There is no single correct way to record your voice. There are many variables and if you manage to reproduce a good sound then fine. However a little common sense and experimentation can pay dividends later, often saving valuable editing time or even having to retake. Here are some points to bear in mind when preparing to make the Podcast.
Being able to recall your setup is an advantage especially when looking for continuity between episodes. Even if you save all of your computer and digital audio settings it will all sound different if you cannot remember the key physical positioning of yourself in relation to the microphone and room. The sound starts at the source, that’s you, the mic and the room. Once you have the optimum positions for the vocal sound make a note of it and add it to your repertoire of production techniques. No amount of adjusting later will correct a poor source signal. So lets have a look at some practical things you can try to improve the recording process.
1) Try the mic in front and parallel to or slightly above your mouth. If the mic is too low you’ll choke your own voice, the throat tends to open with the head tilted slightly back. Also you may need to read from a script which should be positioned in such a way that you don’t turn you head away from the direction of the mic to read it. This will cause a drop in level and a drastic change of frequency or tone. So once you find the best position for overall quality keep your position fixed. Avoid turning pages close to the mic as the noise may be loud and difficult to eliminate later.
Don’t handle the mic as this can cause problems like low frequency rumbling; mount the mic on a stable stand instead. If its table mounted you can place it on top of a good quality mouse mat to help decouple the stand from sound travelling through the table. Ensure the mic/mic stand is isolated from vibrations and noise such as tapping feet, instability or other moving objects nearby.
Try to point the mic away from problem sound sources such as doors, windows, air conditioning and computer noise etc. If needed use a baffle to prevent unwanted sound travelling to the mic, this could be things like duvets, pillows, curtains, carpeting, soft furnishings or board. Remember these materials may also modify the direct/ambient sound entering the mic from the room. It’s usually more practical to go for a "dry" sound rather than a "roomy/boxy/ringy" sound.
The further you are from the mic the more room and background sound will be heard and the more gain will be necessary on the mic input. This can introduce other forms of unwanted noise like distortion and ambient background hiss which can muddy the sound. It’s a good idea to spend some time experimenting with various combinations to hear the effect produced. This usually means doing some trial recordings first. You need to find a happy compromise having taken all things into consideration. Be practical rather than dogmatic.
2) Avoid getting too close to the microphone. This will only introduce a bass tilt known as the "Proximity Effect". This effect is a continual build-up around the 200Hz range which detracts from the clarity of the recording. Unless you’re deliberately going for this effect you’ll find it more practical and safer to maintain a distance of 6 to 8 inches or more. The mic is more sensitive at close range which means even a slight increase in the volume of your voice can become greatly magnified. This can cause many unwanted distortions such as "popping" mouth clicking, breath noise and sibilance (heavy consonants). To feel the effect of this just hold your hand in front of your mouth and vary the distance as you push out some explosive Ps and other sounds.
Putting some distance between you and the microphone allows for small movements of the head off axis which are otherwise very noticeable at close range. The goal is to achieve an even level and balanced tone. Be conscious of your vocal level, too soft will leave a large margin for volume increases whereas a moderate volume is easier to reproduce. The need for a consistent level and tone becomes even more apparent when speech is combined with music. When talking over background sound like music just a slight drop in your vocal level will make it hard to hear what’s being said as the music level remains constant.
One of the most consistent problems associated with being too close to the mike is "popping". This is the name given to explosive bangs that occur when air enters the microphone. You can eliminate it by using a combination of the following:
1) Increase the distance between you and the microphone.
2) Angle the microphone so that you are talking across rather than directly into it.
3) Create a pop shield by forming a wire coat hanger into a circle and stretching a layer of
nylon stocking or tights across it. This can be placed between you and the mic to filter the wind.
4) Carefully tape a pen or pencil to the front of the microphone to block the wind.
Many podcasters like to get close to the mic for a bigger sound with more presence. In practice the reverse can happen for the following reasons:
1) Magnified unwanted noises like sibilance and popping will make it impossible to add EQ and other types of signal processing later. Some sounds will become too harsh when adding treble and pops will become thumps when boosting bass. Adding special effects like reverb or delays will compound the problems further. This can sound much worse when listening to the podcast on headphones. When using headphones to record there is a risk of feedback if the mic is close to an open earpiece. A clean sound at the source can always be made warmer or given presence later far easier and without recourse to corrective processing.
2) If the annoying sounds happen frequently the user will modify the sound accordingly at their end. This can involve anything from changing the tone settings, turning the volume down or even switching it off.
Always check your connections and optimise your record levels with a trial run first. Ensure the levels are not clipping on the loudest parts of your vocal performance. Check that the softer parts are not reading too low as you may run into signal to noise problems with poorer quality equipment. Even up your spoken dynamic range where possible to make the recording safer. When making a recording check for the possibility of interruptions first such as people entering the room or ringing telephones.
Alternative Voice Recording.
In extreme cases it may be impossible to record vocals. If you must have some form of speech in the podcast you can always generate it artificially. Speech synthesis software has become much more sophisticated recently although it still sounds mechanical. However there are some advantages to this technique that in certain circumstances may be preferable. For one the recording process is greatly simplified, requiring no microphone or audio recorder. The user simply types in the required text and generates a digital audio file that can be imported into audio editing software. It is possible to assign a variety of male or female personas for the speech generation and even to translate the text into other languages. Having multiple vocal characteristics allows for the creation of completely different productions, even if it was for demonstration purposes only. Podcasts could be used as transcriptions or multiple language education tools. Perhaps an easily manageable form of text to audio recording could lead to a form of multilingual hyperaudio where anchors in clips and podcasts could be linked between web pages. Effectively sections of audio could be pulled down and appended to other sections to create new audio/video documents and translations. At which point I’ll leave you to imagine the possibilities.
Audio Content And Podcast Usability provides good tips for planning and producing better podcasts.
Podcasting On A Budget talks about basic equipment requirements.
Text To Speech Creator for online voice synthesis.
Recording Vocals provides information about microphones and studio techniques.
SpeechBot tells you about audio search using speech recognition.
Wikipedia Speech Synthesis has all the background info and further links.