Hi All, There are many papers out there on this topic, even some by me from the mid '80s.
I suggest asking the chatbots to find the best papers/methods, then have one (e.g. Claude) write a python version if no code is provided, then finally convert the best code (any language) to Faust (or C++ foreign function if need be). I use Claude Code for writing all software, but also Gemini and ChatGPT for additional literature research (and sometimes Perplexity and Grok). Let us know how the conversion to Faust goes! That's something I've been meaning to explore. Cheers, Julius On Wed, Jan 14, 2026 at 2:36 PM Aaron Krister Johnson <[email protected]> wrote: > Sounds like an interesting problem, but one above my paygrade. > > I think this is Julius O. Smith territory...or one of the other true gurus > here. :) > > Aaron Krister Johnson > Music, etc.: > https://soundcloud.com/aaron-krister-johnson > https://soundcloud.com/filtercreed > https://www.youtube.com/channel/UC_utjGYbSizWE0dNyr0Vdmg > https://aaronkristerjohnson.bandcamp.com/ > http://www.untwelve.org > Code: > https://github.com/akjmicro <http://www.untwelve.org> > > > On Wed, Jan 14, 2026 at 1:34 PM Daz Man <[email protected]> wrote: > >> Hi All, >> >> Does anyone know of a way to write an adaptive allpass filter in Faust? I >> need to be able to adaptively phase shift a pair of signals (perhaps use >> one as a reference and adapt the other) so they never phase cancel at any >> frequency. >> >> I recently wrote some code for an adaptive NLMS FIR filter, and it works >> well for noise reduction, but I have an application where I need to adapt >> phase but not change amplitude. >> >> The purpose is to improve AM demodulation on a signal with multipath >> fading, so the demodulated audio from the two sidebands is always combined >> in-phase so there is never any cancellation. Previously I have tested a >> multiband phase shifter approach for this, but it is not ideal, and it adds >> a lot of group delay. >> >> An alternative might be to attempt to suppress the multipath fading >> notches on the IQ signal before demodulation, but this is far more complex, >> and since the signal modulation is audio and not data there is no training >> signal to base the channel estimate on. >> >> Daz >> _______________________________________________ >> Faudiostream-users mailing list >> [email protected] >> https://lists.sourceforge.net/lists/listinfo/faudiostream-users >> > _______________________________________________ > Faudiostream-users mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/faudiostream-users > -- AI has taken my job, but only I know what it is.
_______________________________________________ Faudiostream-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/faudiostream-users
