The problem with a fixed filter banks for me is experimenting with bandwidths. You have to manually adjust each adjacent band carefully to get the correct effect and that can be cumbersome. I prefer bandpass filters with resonance controls. I use three A-121 Multimode filters to not only get fixed resonances but variable ones like speech. Another way is to use the A-129 Vocoder to automatically simulate acoustic instrument resonances. For example, feed a looped violin or oboe sample from your computer, external synth, A-112, etc. into the vocoder and to resynthesize the spectral shape. --- In Doepfer_a100@yahoogroups.com, "lars_arnwald" <lars.arnwald@...> wrote: > > I'm interested in filters creating formants. It's not about human > voice vowels or the sounds usually produced by by vocoders, but rather > the formants associated with certain acoustic instruments, like cello, > trumpet, trombone, French horn, etc. > Probably nothing can beat the Yamaha FS1R when it comes to parameters > shaping formants, but like many other digitals it can't process > external sounds. I've read the specs for Doepfer A-104 Trautonium > Formant Filter, A-127 VC Resonance Filter, and A-128 Fixed Filter > Bank, as well as Analogue Systems RS-110 Multimode filter, and RS- 360 > Vocal /Phase Filter Bank. Certain details caught my interest like the > input sensitivity with A-104 that easily creates distortion and the > RS-110 external resonance feedback chain where reverbed signals can be > fed back. Some have extensive CV features like A-127, but unless > programmability is desired, this is not a prerequisite since formats > have fixed center frequences, bandwidth, etc, AFAIK. > Unfortunately there are no audio samples for any of the above > mentioned, so I like to ask if any of the list members have looked > into/ experimented with instrument formats. Is there any reading to be > recommended? > I'm new to this list, but it looks quite promising/Lars >
Message
Re: Formant filters
2006-11-05 by laryn91
Attachments
- No local attachments were found for this message.