There is no way to perfectly reconstruct the original sound. You would just reduce the amplitude then try to reconstruct the peaks by interpolating the existing data:īut what if the original waveform wasn't a simple sawtooth? What if the peaks had had a complex shape? Once the peaks have been clipped there is no way to know exactly what they originally looked like. How do you fix these clipped peaks? Looking at the clipped waveform it may be easy to guess it started out as a sawtooth. Those clipped peaks sound very harsh compared to your original source. The peaks of your nice sawtooth wave have now been "clipped" off at 9. The largest value your system can represent is 9. ![]() So all those samples that are two digits need to be reduced to one. Instead of doubling the volume you decide to quadruple it:īut your system only uses single digits at each sample. Instead of ranging from 0 to 4 they range from 0 to 8. It has the same basic sawtooth shape but the peaks are now twice as high. You decide that's not loud enough so you double the volume: That's a simple sawtooth wave that starts at 0k, rises to 4, falls back to 0, and repeats You digitize an audio source and get something like: Say you decide to represent an audio waveform using single digits for the amplitide, 0 to 9.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |