Oq calculation: Apply different methods for distinct portions?

[MinhChauNGUYEN]

For the verification of f0, we can setting different thresholds for several distinct portions.
Regarding Oq calculation, the opening peaks can be also very diverse in one token.
Is that possible if PeakDet has a setting to change the method for a particular portion? And should we do it?

[alexis-michaud]

On principle, yes, it would make sense to detect some of the cycles by means of the pulses in the second derivative, and others by means of a barycentre method applied to the local minima in the first derivative, for instance. Some of this could be automatic or semi-automatic: first, determining for a given token which method yields best results (consistent & clear), then trying to crack the tough nuts: do a diagnosis on the more difficult cycles, to see whether O_q is simply not a relevant measurement for these cycles, or whether a different method gives a clear & convincing result.

This will take serious programming, though. For the sake of your PhD, I'd recommend making do with the current setup, and using just one method for one token. A workaround consists in making manual changes: this gives full freedom to the user to 'pick and choose' from various methods. This complete freedom also allows the user to make big mistakes and to do some unprincipled 'cherry-picking', so it's to be used very carefully, of course.