search

fastlib / fCWT

The fast Continuous Wavelet Transform (fCWT) is a library for fast calculation of CWT.
Apache License 2.0
263 stars 53 forks source link

Can this library do the inverse CWT? #23

Open robclouth opened 1 year ago

robclouth commented 1 year ago

This would be super handy for audio processing

fastlib commented 1 year ago

Not yet, will add it to the list. Are there more people that need this feature?

aphoffmann commented 1 year ago

An ICWT would be very helpful for my research!

filefolder commented 1 year ago

yes would be very handy

fastlib commented 1 year ago

I see there is great interest in a fast inverse CWT. In #36, I mentioned the difficulties with implementing an inverse CWT, however I will put it on my list for the next major update. Thank you for showing your interest.

nicolasvair commented 1 year ago

Yes it would also help me very much !

Side note, do any of you know any sound processing in the wavelet domain ? If existing in python, it would be fontastic to use with neutone AI audio plugin.

InitusNovus commented 10 months ago

It will be very helpful!

My research is ML which makes CWT data(image) If the fast ICWT exists, I will be able to put it in the learning loop, so I will have a chance to step further.

GUIMINLONG commented 9 months ago

inverse fcwt will be very helpful ~~~~l

Sean-M-Devlin commented 7 months ago

Agreed, I'm also interested in performing the inverse too! Thanks for this great tool!

vctrbrtn commented 3 months ago

Likewise, would be interested in the implementation of inverse fCWT.

fastlib commented 1 month ago

I've been tinkering a bit to see whether an inverse fCWT was possible and it is actually very easy! I'm still planning to implement this in the next version, but due to other priorities in my PhD, this update is being postponed constantly.

For now, you can sum all real valued rows of the time frequency matrix (wavelet coefficient matrix):

signal = np.real(tfm).sum(axis=0)

Note that we are not normalizing anything yet, so this is not correct if you are really interested in the absolute amplitudes of the signal. However, if you are planning to normalize the signal anyway, this could be a straightforward solution!