Greetings,

As the title of this thread suggests, I am having amplitude modulation issues when time compressing/expanding with a phase vocoder. Since the time domain signal is windowed both before the FFT and after the inverse FFT, a 4 time overlap is used. This works fine when the analysis and synthesis step sizes are the same, resulting in a constant output gain. However when the steps differ (in order to achieve the time compressions/expansion), the output results in amplitude modulation. The greater the difference between the step sizes, the more obvious this becomes. Surely there is a work around to this, or more likely I am doing something wrong. Do you have any suggestions?

Thanks,

Aristotel

I assume you're doing time compression or expansion by moving your output windows closer together or farther apart (respectively). When you do this, you need fudge your window envelopes.

When you expand, you need to put a hump in the middle of the window envelope to compensate for less overlap. When you compress, you need to dip the middle to compensate for additional overlap.

A bit of geometry and algebra should show you how much and what shape.

Thanks for the suggestion. That would require the synthesis window to be calculate on the fly according to the changes of the synthesis step size, instead of having it precalculated, right? For the time being I have found that moving to a 8x overlap makes the problem go away.

Thanks again!