Is pitch and speed the same in the context of audio programs?

Question

Step means "perceived frequency." Nice. But when I do a very low OpenAL level, the sound plays many times longer. If I do this very high, the sound plays very short, but with a high frequency. For me, the logical consequence is slow or fast.

Or is it a step! = Speed?

Answer 1

Paul R has a pretty good answer, but I would like to expand it a bit. If you think of sound as a series of pulses (and this is how it is), then a higher step will have more pulses per second (higher frequency), and a lower step will be smaller (lower frequency). To lower the pitch of an existing sound, you must propagate these pulses (make them farther apart). As a result, the duration of the sound will increase, because you have not reduced the number of pulses, you just made them even more separate (less per second). The opposite happens if you try to increase the pitch: the pulses are closer to each other, which makes the sound shorter in duration.

If you want the duration to remain constant regardless of changes in the recorded tone, you need to either discard information (lower step) or production information (higher step). This is where fancy processing comes in. What can be safely discarded? What can be safely duplicated or built?

Answer 2

It is relatively easy to vary pitch and rate together, since all you have to do is vary the speed at which you play the samples. If you play samples at half your planned speed, the step will be halved and the sound will come out at half speed (think that the tape deck is operating at the wrong speed). Conversely, playing back samples faster than expected will increase the pitch and make it faster.

You can distinguish pitch and rate independently, but this requires a lot more processing, usually using some kind of analysis and re-synthesis algorithm (e.g. PSOLA for speech).

Answer 3

Using the algorithm that you describe in OpenAL, pitch = f (speed) and speed = f '(pitch).

You can change these parameters somewhat independently using a different algorithm. There are many algorithms for this.

These algorithms can exist in the frequency domain (vocoder, frequency-domain convolution) or in the time domain (PSOLA, WSOLA, extended WSOLA), or both (hybrid models that use time-domain techniques for time partitions and vocoders for tone sections).