I've had an idea for a while that I think will either be completely pointless, or will have some fun implications/applications. What I want to do is invert a sound file.
What I mean by that is that if you were to look at a frequency analysis of any given point of the waveform that it would be inverted. So, using the range of human hearing, anything at 20Hz originally would now be at 20kHz, and vice versa. Or, to put it another way, so that the bass becomes treble and the treble becomes bass.
This is one of those things that I imagine would actually be quite difficult. The way I'd imagine it'd have to be done is that the application would need to use the same calculations used to generate a spectrum analysis for each discrete unit of the sound wave (whatever that would be in this particular example), change the results so that it is inverted, and then generate a new sound wave from that data.
Knowing very little about either programming or the mathematics involved, my question is - how easy would this be to accomplish?
What I mean by that is that if you were to look at a frequency analysis of any given point of the waveform that it would be inverted. So, using the range of human hearing, anything at 20Hz originally would now be at 20kHz, and vice versa. Or, to put it another way, so that the bass becomes treble and the treble becomes bass.
This is one of those things that I imagine would actually be quite difficult. The way I'd imagine it'd have to be done is that the application would need to use the same calculations used to generate a spectrum analysis for each discrete unit of the sound wave (whatever that would be in this particular example), change the results so that it is inverted, and then generate a new sound wave from that data.
Knowing very little about either programming or the mathematics involved, my question is - how easy would this be to accomplish?
via International Skeptics Forum http://ift.tt/29krOLw
Aucun commentaire:
Enregistrer un commentaire