What is "Conceptual Representation" of sounds?

bjkwon
Posts: 111
Joined: September 26th, 2018, 9:36 pm

What is "Conceptual Representation" of sounds?

When writing AUX codes, the user should consider conceptual factors describing sounds and sound processing rather than their precise technical specifications. For example, a mathematical description of a pure tone of f Hz for a duration of d sec is

sin(2πft), where 0<t<d.

And its technical description for implementation is a precise representation of digitized samples according to the equation. We need to consider an actual digital system to proceed; i.e., need to settle on the sampling rate of the digital system.

Then, actual coding, for an example of a sampling rate of fs Hz and a duration of d sec, as follows (this is in MATLAB but the principle is universal):

fs=22050;
t=(1:fs*d)/fs;
t=sin(2*pi*f*t);

Now, the philosophy of AUX is, to relieve the users from the burden of implementation as much as possible. The question to ask ourselves is, are you an "audio" person or a digital system engineer including software developer responsible for audio device drivers? If you are the latter, forget AUX; but if you are the former, you really shouldn't worry too much about the details of implementation. For you, what's important is, I want to have a tone with a frequency of f Hz for a duration of d Hz. I want everything else taken care of under the hood. So here comes the AUX code---

tone(f,d)

This does not mean that audio people don't need to know anything related to implementation. There are a number of reasons for everyone to have a solid understanding of digital signal processing. This is not just a lip service. There are crucial needs in the practical sense (see below). However, if you work on the sound itself, you would often find details on implementation techniques tedious and cumbersome.