Additive Synth Tools
2003-05-06 by paulhaneberg
I have been playing around a lot with additive synthesis on the MOTM using a number of oscillators. I am wondering if anyone has any suggestions for a piece of software capable of doing analysis of a single note played by a particular instrument, for example a wav file of a sax. I am looking for something that can give me a time varying spectra so that I can see the individual envelopes of each harmonic as well as determine the individual harmonic amplitudes. I have been using SpectraFoo inside my ProTools setup. Although it works rather well for giving me an average relationship between harmonics, it does not show the time varying relationship. Also helpful would be if anyone knows of a source (book or website) for this sort of information in a quantitative form, not just some general looking graphs. I have also been contemplating trying to build a MOTM style Fourier series generator. This is a big design project, so I'm asking if anyone has any ideas to let me know. I'd rather not reinvent the wheel. I do realize I can do this sort of thing with a lot of MOTM oscillators (as I did on the ELP thing) but I'd like to find something a little lower in cost than buying 16 MOTM300s ;) And it's not easy to tune a number of 300s precisely enough to get a good harmonic series even using sync. (I've found it is difficult to get sync once you get above the 8th harmonic, at least on certain partials.) My idea is to take the triangle output of a MOTM300 and frequency multiply it using a series of rectifiers and amplifiers up to perhaps 256 times the original frequency. (A rectified triangle wave is a triangle wave at twice the frequency of the original.) The resulting waveform would be transformed into a clock signal driving a series of counters which would count from 0 to 255 then reset. Each counter would address a sine wave lookup table with a single sine wave divided into 256 increments. Each counter would increment by an amount equal to the harmonic it was generating. For instance the 3rd harmonic counter would increment by three on each pulse, the 13th harmonic counter would increment by 13 each pulse. This would create a series of sine waves albeit digitally produced which would then be filtered down so that there would be individual outputs for each harmonic. Presumably the whole thing would be multiplexed and perhaps implemented within a PIC if it's capable of going fast enough. The harmonics above the audio range would not necessarily need to be reproduced so perhaps the original frequency could be detected and used as a means of determining which harmonics were too high to be useful and not reproduced thus saving overhead on the processor. Each output would have its own attenuator, and ideally each would be fed into a VCA with and EG, I think a VCF driven by and EG could substitute as upper harmonics tend to decay faster than lower ones almost without exception. I would think such a module would be very useful and interesting to play around with although I'm sure someone will suggest I get a Kawai K5000. (I have one.)