[Elecraft] optimizing recorded audio

[Ian White GM3SEK]

Jim Brown wrote:
>> Finally, go to "change tempo." This can make your recording play faster or
>> slower without changing pitch. Don't change the beats, but rather trim the
>> time length of your recording. A little bit goes a long way. For example,
>> when I recorded my call the clip was about 3 seconds long. Cutting it to 2.6
>> seconds got rid of inter-syllabic pauses etc and made it sound urgent but
>> didn't introduce any unnatural sound.
>
>Time compression tends to make voices sound artificial. Time 
>compression, or talking too fast with poor articulation, can make it 
>difficult for others to copy your call.

That is confusing a problem with its solution. The problem, everyone 
agrees, is trying to talk too fast with poor articulation. But time 
compression isn't part of that problem - applied correctly, it is part 
of the CURE.

Try this: record your messages (CQ and callsign) focusing 100% of your 
attention on speaking clearly. Don't worry at all about speed; speak as 
slowly as you find necessary to get good articulation. Then you can use 
the 'tempo' function to bring the recording up to a normal speed.

Most people can improve articulation dramatically by slowing down only 
10-20%, so it only requires a modest increase in the tempo setting to 
restore a normal brisk speed. Time compression is a re-sampling 
technique and it does introduce some artefacts, but these are minor 
compared with everything else that happens to a SSB voice signal.

At this point, you can bring in a third factor: pacing. If the gaps 
between words or syllables don't sound quite right, you can experiment 
by cutting out (or pasting in) small segments of 'quiet time'. This is a 
very simple cut-copy-paste operation in Audacity, easy to follow on the 
scope trace.

The golden rule is: listen to the results after every step in the 
editing process. If it doesn't sound good, then Undo that step and try 
something else.

If it's done well (which really isn't hard, and quite fun to learn), 
you'll find that your voice sounds clearer, but still quite natural. In 
fact, you'll probably sound *more* natural than if you were straining 
for optimum articulation, speed and pacing, all at the same time. And 
after 24-48 hours of SSB contesting, your recorded voice is *guaranteed* 
to sound better than your natural self!


-- 

73 from Ian GM3SEK
http://www.ifwtech.co.uk/g3sek