Analyzes a first order ambisonic signal (B-format).
The input signal, an array: [in0, in1, ... inN]
NOTE: All analyzers take args of
Arguments (listed above with each 'kind') for the wrapped transformer UGens. Arguments can NOT be passed in through keyword through the FoaAnalyze wrapper. You can pass values in by keyword if you use the transform UGens directly.
Analyzers offer two methods of analysis:
These keywords describe the form of the underlying filter method employed to return the various analyses. Instantaneous time domain soundfield analysis is offered by
'instant' where time averaged time domain analysis is returned by
The instantaneous analysis method uses Hilbert Transform Filters to generate an Analytic Signal from which the resulting time domain analysis is derived. As a result, the
'instant' analysis method is able to return a detailed view of both the active and reactive soundfield components.1 Additionally, as implied by the name,
'instant' returns the instantaneous sample-by-sample analysis of soundfield features. As you would expect, the resolution of the analysis is determined by the resolution of the underlying Hilbert Transform Filters. Increasing the size argument to larger values increases the frequency domain bandwidth of the analysis. The Hilbert Transform Filters employed are linear phase for the Real response, and therefore introduce a delay. Since the Fast Fourier Transform is used to realize the Hilbert Transform Filters, size is constrained to a power of 2.
The time averaged analysis method,
'average', is implemented through the use of an un-weighted Moving Average Filter. This filter also has a linear phase response, and therefore the time averaged method also introduces a delay. Rather than a sample-by-sample view, this analysis offers an efficient time averaged reading of the soundfield. While detailed inspection of the active features of the soundfield is available, some aspects of the reactive soundfield are invisible to the time average method.2 As with
'instant', increasing the size argument to larger values increases the bandwidth of the analysis. The size of the Moving Average Filter is not constrained to a power of 2.
These features are summarized:
|instantaneous time||Active & Reactive Vectors||Hilbert Transform||higher|
|averaged time||Active Vectors, Reactive Magnitudes||Moving Average||lower|