Extra Frequencies when speed is changed.

[Marc Hildebrant]

Not sure if it's a bug or not :

If you created a 1000 Hz tone and then change the speed of the waveform using the "change speed" effect, you create the correct new tone around 1360 Hz, BUT, you also create tones around 15 KHz.

Marc

[Craig Maier]

Hi,

What dB level does this new signal occur at? Is it down in the mud? Does the same thing happen with files that have higher sampling rates such as 96 kHz?

Craig

[Craig Maier]

I created a 1 kHz sine wave at -10 dB (mono) and 44.1 kHz and 16 bit depth. I changed the speed by +35 %. I measured a spurious signal at 17 kHz that was 66 dB below the 1335 Hz fundamental. Next, I created another 1 kHz sine wave at -10 dB (mono) and 96 kHz and 16 bit depth. I changed the speed by the same amount (+ 35 %). I measured a spurious signal at 17 kHz and that was 99 dB below the fundamental. (The shifted sine wave frequency fundamental became 1335 Hz due to the speed changer.)

I attribute these spurious signals as artifacts of the non-ideal anti-aliasing filters used as part of the algorithm. Of course, the higher the sample rate, the farther away we get from the corner frequency of the filter and thus the lower the level of the spurious signal.

In either case, a 17 kHz signal that is 66 dB or 99 dB below the source frequency of 1335 Hz is not audible by humans. So, it should not be a practical problem.

Craig

[Marc Hildebrant]

Craig,

Thanks for checking it out. Indeed the level is very low and should not be a problem.

Marc