I am working on an audio analysis project in C++ and wish to use the sswt, but I'm not sure how to optimize the threshold for the division. If the threshold is too strict, it cuts off the higher frequencies (lower amplitudes), and not strict enough, the floating point errors get amplified.
I am working on an audio analysis project in C++ and wish to use the sswt, but I'm not sure how to optimize the threshold for the division. If the threshold is too strict, it cuts off the higher frequencies (lower amplitudes), and not strict enough, the floating point errors get amplified.