We recently attended some listening tests hosted by NGW to demonstrate a new codec for producing MPEG-2 audio for DAB, We also had the chance to listen to some DAB+ audio.
It was a very enjoyable experience although I admit I have never been very good at listening tests. I can’t always distinguish subtle difference in audio as easily as the eagle eared engineers, in part due to the fact that my hearing isn’t very sharp. In any case we have to be very careful when discussing subjective opinions about audio.
Listening to the audio from the different codecs, however, I could tell the difference – just. One was more lean and rangy – more treble, basically.
Now here is why subjective listening tests are to be viewed with caution. If you’re turning out jangly guitar rock you might quite like the sound quite lean and rangy. If however you’re running a talk service or playing a lot of acoustic numbers then that lean and rangy treble might be less welcome.
In addition, whatever we provide in terms of baseline parameters can be totally masked by a station’s processing – or lack of, and the quality of the same. There is excellent and disastrous unprocessed audio and the same goes for various types of processing. How a radio station presents its audio is very much up to them. So for us as multiplex operators, just picking one codec over another alone does not guarantee a subjective improvement for you the listener.
Luckily there is hard data we can refer to – data that demonstrates the actual frequency range that each codec achieves. A greater range gives services a broader platform to stand on. So subjective listening is valuable but is only part of the decision making process.