CrossSynthesis(RE,EX) - Kyma Forum

Tips & Techniques

CrossSynthesis(RE,EX)

Post New Topic Post A Reply
profile | register | preferences | faq | search

next newest topic | next oldest topic

Author	Topic: CrossSynthesis(RE,EX)
marinos g Member	posted 11 May 2007 07:21 Hello everybody, can i have a collective input on how to treat my sounds with this analysis without having unstable output? basically a short explanation on the type of sounds to be used and parameter explanation.a basic tutorial.the result seems to me much more interesting than the crossfilter.the characteristics of the sounds used for exiter and resonator seem to be more 'present' at the resultant sound and that is why i prefer that type of analysis-synthesis for the creation of hybrid sounds.But... it seems to be very sensitive procedure that if not done correctly produces lots of artifacts. IP: Logged
SSC Administrator	posted 11 May 2007 08:26 As a general guideline, listen to the EX file by itself; it is very low amplitude and the frequency spectrum is broadband. The best inputs to an RE filter are, like the EX files, low amplitude and broadband (like noise, pulse trains, sawtooth waves, etc). The RE filter is all poles so it can blow up very easily unless the amplitud of the input is extremely low. Some experiments you can try: * Do the analysis on high-pass filtered versions of the original (to de-emphasize the fundamental and concentrate on the formants) * Compress the input to the RE and the RE itself to ensure that there is always some sound coming through the filter and to avoid too much dynamic range on the input Not suggesting to do all of the above at once--these are just things to experiment with. IP: Logged
marinos g Member	posted 15 May 2007 11:50 yes but is it possible to explain a little bit what happens during the analysis and what the parameters on the analysis tool like 'averaging time 'mean? IP: Logged
SSC Administrator	posted 15 May 2007 12:22 On each "frame", the analysis tries to match the spectrum by positioning the poles of the filter. When you place a pole at a certain frequency, the filter will emphasize that frequency. The more poles you use (the higher the Filter Order), the closer you would get to a pole at each partial of the spectrum. However, we are looking for a spectral envelope not an exact match of the spectrum, so it is advisable to use fewer poles than you would use, say, oscillators in a SumOfSines resynthesis. (Similarly when you look for the amplitude envelope of a signal, you do not want each and every sample; you want a kind of running average of the samples squared so they are all positive values). Averaging time is the duration of the sample that is analyzed per frame. There is more info on the RE starting on page 196 of Kyma X Revealed. IP: Logged

All times are CT (US)	next newest topic \| next oldest topic
Administrative Options: Close Topic \| Archive/Move \| Delete Topic

Contact Us | Symbolic Sound Home

This forum is provided solely for the support and edification of the customers of Symbolic Sound Corporation.

Ultimate Bulletin Board 5.45c