I'm aware that this topic doesn't belong into this forum section, but as it was already mentioned, it needs to be clarified so someone wouldn't make wrong conclusions.
CD audio quality is defined by:
*sample rate of 44.100kHz -that's the lowest sample rate at which flat frequency response of 20Hz-20kHz is possible (in stereo).
*bit depth of 16bit -defines sampling precision of dynamic range (over-simplified: precision of loudness).
*wav format -is recorded data stream (means, series of 16bit numbers sampled at given sample rate) saved on media.
-that's the shortest simplified definition.
If we look at keyboard specifications, we can see, that the sound from our keyboard has exactly the same specifications. And it doesn't matter if we record digitally on keyboard, or if we use the audio output on back panel -because the source is the same.
..But, of course, you can record/listen at a higher bit rate i.e. 24-bit @192kHz, etc...
-yes, one can do that, but the quality will still be the same (as coming from keyboard). In short: quality of the sound depends on source. We can't change that afterwards by increasing recording parameters.
Higher bit- and sample rates are (only) used for master recordings in studios. Reason being, to make final product to sound as intended, a lot of sound post-processing is needed. And by doing this, audio quality might degrade. Means, master track of higher quality is used, so final sound still can have CD quality.
Now, depending on compression settings, mp3 still has sample rate of 44.100kHz and bit depth of 16bit. So where's the catch? It's called
bit rate, which can be 8kbps all the way up to 320kbps. Chosen bit rate defines amount of compression, which directly influences the size of the file.. and the quality of sound.
So what is actually sacrificed in mp3? The shortest answer would be: frequency response. But in reality, it's not that simple. Depending on compression settings, only some (not all) of high frequencies are removed. This can easily be seen (on spectrogram) if we choose bit rate of 192kbps. In this case, we can see, that sound between 16kHz and 18kHz is still present. However not all frequencies in this range are covered. Which frequencies will remain depends on source (music complexity). Anyway, mp3 doesn't cut at low frequency range. The thing is, while we can hear (and "feel") low frequencies regardless of our age, high frequency perception degrades all the time as we age -without exception. And so, majority of people above the age of 30 can't hear sounds much above 16kHz. Yeah, guys at Fraunhofer Institute were quite smart when they invented mp3.
But one thing needs to be mentioned. With wav file, audio quality doesn't degrade if we re-save it again and again. That's not necessary the case for mp3 files -at the end, it's called "lossy" format. And finally, not all mp3 codec behave exactly the same. And so, for archiving purpose (of music we make on keyboard), I recommend using wav file, which we can use as master file in case of later post-processing. Or even better, we convert wav to flac format -that way we reduce file sizes substantially and still keep the original quality.
Bogdan