how audio is converted to audio data?

Sound is made of pressure waves. A microphone translates the pressure waves into an electric signal composed by (almost) the same 'shape', this time what change is voltage and not air pressure. Then the electric signal passes throught a ADC (Analog to Digital Converter - http://en.wikipedia.org/wiki/Analog-to-digital_con... Wikipedia tells you all about the electric signal changing all the time while you have to choose the rate at wich values are sampled, the higher the sampling rate, the better the samples quality and the subsequent reconstruction of the electric signal from the samples.

But you have not to overdo. There's a theorem that tells you which sampling rate is 'enought' and the rule of thumb resulting is 'the sample rate should be at least twice the higher frequency you want to sample'.

With the voice that passes throught a telephone, 8000 samples/seconds is enought - the telephone line has already 'mangled' the sound, so that a better sampling is useless (tech: the telephone cuts all the frequencies above 4 kilohertz).

When you want to sample all the frequencies an human ear can feel, you need to sample correctly frequencies at 22 kilohertz, therefore the sampling rate has to be 44000 samples/second (CD quality) or something more (DAT quality).

http://en.wikipedia.org/wiki/Analog-to-digital_con...

"Mr Watson—Come here—I want to see you"

...even earlier

http://en.wikipedia.org/wiki/Alexander_Graham_Bell...