|VK3OTN Broadcast Article Recording & Production
IIn order to produce 11 monthly news bulletins each year, here are some tips on how to go about it
Both VK1WIA and VK6ARN produce a weekly half hour news and information
bullitin, of a high technical quality, which is available as an mp3
file. It is at a 64 Kbps bit rate and a 44.1 kHz sample rate, and
around 13 MB in size. VK3OTN transmit a monthly news bulletin, with a
focus on club news and items of historical interest, relating to radio
and electronics, to be transmitted eleven times a year on the first
Monday of each month, except January. Both the previous mentioned
broadcasts are usually of a high quality, and there's no reason,
provided a little care is taken, that VK3OTN should not achieve the
same degree of professionalism.
The goal is to produce audio of a gold standard. Once an audio signal
is degraded with noise, distortion, and other artefacts that make it
sound akin to a poor cellular telephone call, it's very difficult to
turn ‘lead back into gold’. Compressing audio files and then
uncompressing them and repeating this several times, can quickly
degrade the audio quality. It's best not to do this in the first place.
The mp3 audio file that is distributed to stations transmitting the
monthly VK3OTN broadcasts is the same file as can be downloaded from
the web site on the weekend prior to the monthly broadcast.
Here are some tips, in no particular order, on how to go about producing a good quality recording.
- Audio should be recorded as a 16 bit 44.1 kHz mono,
linear PCM file, using no software compression whatsoever when saving
the file. This is effectively the same quality as used by audio compact
discs. CD's, however, are actually 14 bit stereo, linear PCM.
- Ensure the audio being recorded is less than the peak
input level of the peak reading meter, otherwise distortion will be
introduced. Softer sounds can easily be fixed up with software.
- Audio should then be normalised, that is the peak of the signal is plus and minus unity, or full scale.
- It is possible to use software to lift the perceived
loudness of the recorded audio. Usually 3 or 6 dB is all that's needed,
The peaks will be clipped, however the signal will suffer minimal
distortion and the perceived loudness should be improved. It should
- An external VU meter, or better still a software VU
meter can give a good visual indication as to the loudness. Just
because the audio is normalised on the various segments that combine to
produce the broadcast, it doesn't mean it will all sound the
same. One well known example is LOUD TV commercials. The audio is at
the same peak level as the program being broadcast, however a few
simple tricks are used to make it sound louder in order to make you sit
up and pay attention. The audio of the entire broadcast should have the
same level, or perception of loudness, as heard by the human ear.
- Ensure the room the audio is being recorded is
acoustically quiet. No dogs, magpies, budgerigars - or other noisy pets
for that matter – or rain on the tin roof; also don't forget to silence
or move any ticking clocks
- Use a quality microphone. A combined microphone and
headphone 'headset' is one option. This way the microphone is a
constant distance from the mouth. Desktop condenser microphones start
at around $25 or so. A USB microphone, which plugs directly into the PC
is one option. The A to D conversion takes place right at the
microphone so it is not possible for electrical noise to be introduced.
A windsock on whatever microphone is used is also a good idea to reduce
breathing sounds and other artefacts.
- Prepare what you are going to read on a word processor
by typing it out precisely how you intend to say it. You can either
print it out, or read it directly off the screen. Ensure the font size
is nice and big, thus making it easy to read.
- Very experienced readers (an example is professional
news readers we hear and see on broadcast radio and television) can
read a script faultlessly. This only comes about from years of
practice. Unfortunately, most of us don't have that experience and will
stumble and make mistakes. Umms, errs and stuttering are common errors.
Don't strive for perfection and don't stop recording. Simply leave a
short pause and start again at the beginning of the sentence. The
mistakes can be easily removed with software later on. If you're
reading an article or story, do the recording in one continuous
session. Don't attempt to do one article using several different
recording sessions, hoping to join the two or more recordings together.
When you join the various segments together, a difference will be noted.
- If sending a segment to whoever's producing the
finished broadcast file via email, and if the file is greater than four
minutes in length, around 20 MB, then use mp3 compression. Use a 256
Kbps bit rate, and leave the sample rate alone at 44.1 kHZdio
There's many different program suites available, some free and some at a cost, for the three main desktop operating systems.
• Audacity (free).
• NCH Wavepad ($).
• Syntrillium Cooledit (bought by Adobe and rebranded Audition).
• Adobe Audition ($).
• Software VU meter (free). A web search for 'vumeter james chapman' will find something quite suitable
The RAOTC broadcast team look forward to hearing from you if you are
able to assist in any of these activities. Many hands make light work!