Arithmetic Operations

Arithmetic operations can be applied in the time and frequency domain and are implemented in the methods add, subtract, multiply, divide, power and matrix_multiplication. For example, two Signal, TimeData, or FrequencyData instances can be added in the time domain by

>>> result = pf.add((signal_1, signal_2), 'time')

and in the frequency domain by

>>> result = pf.add((signal_1, signal_2), 'freq')

Note that frequency domain operations are performed on the raw spectrum signal.freq_raw.

Arithmetic operations also work with more than two instances and support array likes and scalar values, e.g.,

>>> result = pf.add((signal_1, 1), 'time')

In this case the scalar 1 is broadcasted, i.e., it is is added to every sample of signal (or every bin in case of a frequency domain operation). The shape of arrays need to match the cshape of the resulting audio object, e.g.,

>>> x = np.arange(2 * 3 * 4).reshape((2, 3, 4))
>>> y = pf.Signal(np.ones((2, 3, 4, 10)), 44100)
>>> z = pf.add((x, y))

or are broadcasted, e.g.,

>>> x = np.arange(3 * 4).reshape((3, 4))
>>> y = pf.Signal(np.ones((2, 3, 4, 10)), 44100)
>>> z = pf.add((x, y))

where y is a signal with y.cshape = (2, 3, 4) and a length of 10 samples.

The operators +, -, *, /, ** and @ are overloaded for convenience. Note, however, that their behavior depends on the Audio object. Frequency domain operations are applied for Signal and FrequencyData objects, i.e,

>>> result = signal1 + signal2

is equivalent to

>>> result = pf.add((signal1, signal2), 'freq')

and time domain operations are applied for TimeData objects, i.e.,

>>> result = time_data_1 + time_data_2

is equivalent to

>>> result = pf.add((time_data_1, time_data_2), 'time')

In addition to the arithmetic operations, the equality operator is overloaded to allow comparisons

>>> signal_1 == signal_2

See audio classes for a complete documentation.

FFT Normalizations

The arithmetic operations are implemented in a way that only physically meaningful operations are allowed with respect to the FFT normalization. These rules are motivated by the fact that the normalizations correspond to specific types of signals (e.g., energy signals, pure tone signals, stochastic broadband signals). While addition and subtraction are equivalent in the time and frequency domain, this is not the case for multiplication and division. Nevertheless, the same rules apply regardless of the domain for convenience:

Addition, subtraction, multiplication, and power

  • If one signal has the FFT normalization 'none', the results gets the normalization of the other signal.

  • If both signals have the same FFT normalization, the results gets the same normalization.

  • Other combinations raise an error.

Division

  • If the denominator signal has the FFT normalization 'none', the result gets the normalization of the numerator signal.

  • If both signals have the same FFT normalization, the results gets the normalization 'none'.

  • Other combinations raise an error.