Vocoders use a type of cross synthesis. Cross synthesis involves using the parameters of one sound to synthesize or process similar parameters in another sound. In the case of a vocoder, the amplitudes of frequency bands of a modulation audio signal are used to set the gain levels for bandpass filters that are applied to a carrier audio input signal. Frequency bands of the modulator signal which have little or no amplitude result in cutting or eliminating those frequency bands in the carrier signal. Frequency bands of the modulator signal that have strong amplitudes lead to boosts in gain for those frequencies if present in the carrier signal.
Vocoding in Reason is accomplished with the BV512 Vocoder. Since vocoders require two audio devices to work, you will need to create a two different synthesizers and /or samplers to connect to the unit. Generally speaking, your carrier signal is harmonic, frequency rich, and rather consistent in spectral content over time. A sawtooth or pulse wave (with short duty cycle) are classic examples of good carrier signals. Your modulation signal should have a changing spectrum, which will provide dynamic (changing) filter parameters over time. Using speech as a modulator with a pulse or saw wave as carriers creates the classic talking robot sound.
Modulating signals must be monophonic for the BV512. You can accomplish with either a line mixer or the Spider Audio Merger and Splitter. If you use the merger, plug both outputs from your mod device into the left inputs of the merger (not left and right, because that doesn’t result in a merged signal). Follow the flow chart on the back of the device. If you use a line mixer, you can control individual levels, and can even use multiple synths combined with level control as a single mod signal.
Since you will most likely need to trigger the carrier and the modulator for every note, you should use a Combinator device. If you already have your carrier, mod, merger, and vocoder devices in the rack, you can select all of them (shift-click on each) and choose “combine” from the edit window (or right-click and choose from the contextual menu). If you have nothing, start with the combinator, and create your devices inside it. All the devices will respond to the same MIDI gate if you assign a MIDI channel to Combinator – MIDI in the Advanced Control section of the interface. The output of the vocoder should connect to the input of the combinator, and the output of the combinator should go to the hardware interface. Save your patches for your indivdual audio devices within those devices, but save a patch for the combinator. This will save your vocoder settings, and all your internal routing within the combinator.
The vocoder controls are relatively simple. You choose the number of frequency bands you want to use (for filters). 512 bands was not common on original vocoders. This band setting is more of a spectral cross synthesis, and gives you the most mapping of one sound to another. 32 bands is the more common setting. At 16, 8, and 4 bands you lose some of the individuality of the modulator, but this doesn’t mean it’s wrong to use these settings. In fact, changing between low and high band values can create some interesting variations with the same signals. In the filter band display you will see how the filters are being set according the modulation signal. Below those constantly changing leds are sliders that let adjust the individual frequency bands. Adjustments can help with distortion when modulator and carrier have a lot of energy in the same frequency bands. The bottom right of the unit has a wet/dry control. Wet means the output signal is the result of the processing. Dry in this unit’s case means you hear the modulator signal only.
The envelope settings can be very useful for creating musical variation. The envelope applies to how long the frequency gain settings hold their places. Short attack and decay times mean that the filter settings will fairly closely track the input modulator signal characteristics. Using longer decay times means that the filter settings will hold longer, skipping over some of the input mod signal. The filter settings don’t update until the envelope has finished its cycle. A longer decay setting creates a slower, sliding amplitude changes. Again, this effect allows for variation using the same mod and carrier signals.
Your carrier and/or modulator can be sequenced using the Matrix or the step sequencer in the Thor synth. You can use a Redrum for either source as well. I’ll try to set up some examples for next class.