voice_opt: Voicing options
In pmcharrison/voicer: Analyzing and Generating Chord Voicings

Defines a list of options to be passed to voice.

voice_opt(min_octave = -2L, max_octave = 1L, dbl_change = TRUE,
  min_notes = 1L, max_notes = 4L, features = voice_features(),
  weights = voice_default_weights, exp_cost = FALSE,
  norm_cost = FALSE, log_cost = FALSE, verbose = interactive())

`min_octave`	(Integerish scalar) The minimum octave allowed in the voicing, expressed relative to middle C (MIDI note 60). For example, -1 identifies the octave ranging from one octave below middle C to the B below middle C.
`max_octave`	(Integerish scalar) The maximum octave allowed in the voicing, expressed relative to middle C (MIDI note 60). For example, 0 identifies the octave ranging from middle C to the B 11 semitones above.
`dbl_change`	(Logical scalar) Is it permitted to change the doubling of the chords, whether by adding duplicated pitch classes or removing duplicated pitch classes?
`min_notes`	(Integerish scalar) Sets the minimum allowed number of notes in the voiced chords; ignored if `dbl_change = FALSE`.
`max_notes`	(Integerish scalar) Sets the maximum allowed number of notes in the voiced chords; ignored if `dbl_change = FALSE`.
`features`	A named list of features to apply to chord transitions, defaulting to a list created by `voice_features`. Any new features must be created using `cost_fun`, and take chords as arguments, where the chords are represented as ordered numeric vectors of MIDI note numbers.
`weights`	A numeric vector of weights mapping to the respective elements of `features` in their original order. This vector need not be named, but an error will be thrown if names are provided and they do not match with those of `features`. Alternatively, it is possible to pass a voicer model as created by `model_features`, or the `weights` slot of such a model.
`exp_cost`	(Logical scalar) Experimental - passed to `seq_opt`, thereby determining whether the linear predictor is exponentiated when computing a sequence's cost.
`norm_cost`	(Logical scalar) Experimental - passed to `seq_opt`, thereby determining whether the (possibly exponentiated) linear predictor is normalized over all possible continuations when computing a sequence's cost.
`log_cost`	(Logical scalar) Experimental - passed to `seq_opt`, thereby determining whether to take the logarithm of the (possibly exponentiated, possibly normalized) linear predictor when computing a sequence's cost.
`verbose`	(Logical scalar) Determines whether progress messages are printed during the function's execution.