Main Content

spectralFlatness

Spectral flatness for audio signals and auditory spectrograms

Description

example

flatness = spectralFlatness(x,f) returns the spectral flatness of the signal, x, over time. How the function interprets x depends on the shape of f.

example

flatness = spectralFlatness(x,f,Name=Value) specifies options using one or more name-value arguments.

[flatness,arithmeticMean,geometricMean] = spectralFlatness(___) returns the spectral arithmetic mean and spectral geometric mean. You can specify an input combination from any of the previous syntaxes.

example

spectralFlatness(___) with no output arguments plots the spectral flatness.

  • If the input is in the time domain, the spectral flatness is plotted against time.

  • If the input is in the frequency domain, the spectral flatness is plotted against frame number.

Examples

collapse all

Read in an audio file and calculate the flatness using default parameters.

[audioIn,fs] = audioread("Counting-16-44p1-mono-15secs.wav");
flatness = spectralFlatness(audioIn,fs);

Plot the spectral flatness against time.

spectralFlatness(audioIn,fs)

Figure contains an axes object. The axes object contains an object of type line.

Read in an audio file and then calculate the mel spectrogram using the melSpectrogram function.

[audioIn,fs] = audioread("Counting-16-44p1-mono-15secs.wav");
[s,cf,t] = melSpectrogram(audioIn,fs);

Calculate the flatness of the mel spectrogram over time.

flatness = spectralFlatness(s,cf);

Plot the spectral flatness against the frame number.

spectralFlatness(s,cf)

Figure contains an axes object. The axes object contains an object of type line.

Read in an audio file.

[audioIn,fs] = audioread("Counting-16-44p1-mono-15secs.wav");

Calculate the flatness of the power spectrum over time. Calculate the flatness for 50 ms Hamming windows of data with 25 ms overlap. Use the range from 62.5 Hz to fs/2 for the flatness calculation.

flatness = spectralFlatness(audioIn,fs, ...
                            Window=hamming(round(0.05*fs)), ...
                            OverlapLength=round(0.025*fs), ...
                            Range=[62.5,fs/2]);                

Plot the spectral flatness.

spectralFlatness(audioIn,fs, ...
                 Window=hamming(round(0.05*fs)), ...
                 OverlapLength=round(0.025*fs), ...
                 Range=[62.5,fs/2]);

Figure contains an axes object. The axes object contains an object of type line.

Create a dsp.AudioFileReader object to read in audio data frame-by-frame. Create a dsp.SignalSink to log the spectral flatness calculation.

fileReader = dsp.AudioFileReader('Counting-16-44p1-mono-15secs.wav');
logger = dsp.SignalSink;

In an audio stream loop:

  1. Read in a frame of audio data.

  2. Calculate the spectral flatness for the frame of audio.

  3. Log the spectral flatness for later plotting.

To calculate the spectral flatness for only a given input frame, specify a window with the same number of samples as the input, and set the overlap length to zero. Plot the logged data.

win = hamming(fileReader.SamplesPerFrame);
while ~isDone(fileReader)
    audioIn = fileReader();
    flatness = spectralFlatness(audioIn,fileReader.SampleRate, ...
                               'Window',win, ...
                               'OverlapLength',0);
    logger(flatness)
end

plot(logger.Buffer)
ylabel('Flatness')

Figure contains an axes object. The axes object contains an object of type line.

Use dsp.AsyncBuffer if

  • The input to your audio stream loop has a variable samples-per-frame.

  • The input to your audio stream loop has an inconsistent samples-per-frame with the analysis window of spectralFlatness.

  • You want to calculate the spectral flatness for overlapped data.

Create a dsp.AsyncBuffer object, reset the logger, and release the file reader.

buff = dsp.AsyncBuffer;
reset(logger)
release(fileReader)

Specify that the spectral flatness is calculated for 50 ms frames with a 25 ms overlap.

fs = fileReader.SampleRate;

samplesPerFrame = round(fs*0.05);
samplesOverlap = round(fs*0.025);

samplesPerHop = samplesPerFrame - samplesOverlap;

win = hamming(samplesPerFrame);

while ~isDone(fileReader)
    audioIn = fileReader();
    write(buff,audioIn);
    
    while buff.NumUnreadSamples >= samplesPerHop
        audioBuffered = read(buff,samplesPerFrame,samplesOverlap);
        
        flatness = spectralFlatness(audioBuffered,fs, ...
                                   'Window',win, ...
                                   'OverlapLength',0);
        logger(flatness)
    end
    
end
release(fileReader)

Plot the logged data.

plot(logger.Buffer)
ylabel('Flatness')

Figure contains an axes object. The axes object contains an object of type line.

Input Arguments

collapse all

Input signal, specified as a vector, matrix, or 3-D array. How the function interprets x depends on the shape of f.

Data Types: single | double

Sample rate or frequency vector in Hz, specified as a scalar or vector, respectively. How the function interprets x depends on the shape of f:

  • If f is a scalar, x is interpreted as a time-domain signal, and f is interpreted as the sample rate. In this case, x must be a real vector or matrix. If x is specified as a matrix, the columns are interpreted as individual channels.

  • If f is a vector, x is interpreted as a frequency-domain signal, and f is interpreted as the frequencies, in Hz, corresponding to the rows of x. In this case, x must be a real L-by-M-by-N array, where L is the number of spectral values at given frequencies of f, M is the number of individual spectra, and N is the number of channels.

  • The number of rows of x, L, must be equal to the number of elements of f.

Data Types: single | double

Name-Value Arguments

Specify optional pairs of arguments as Name1=Value1,...,NameN=ValueN, where Name is the argument name and Value is the corresponding value. Name-value arguments must appear after other arguments, but the order of the pairs does not matter.

Before R2021a, use commas to separate each name and value, and enclose Name in quotes.

Example: Window=hamming(256)

Note

The following name-value arguments apply if x is a time-domain signal. If x is a frequency-domain signal, name-value arguments are ignored.

Window applied in the time domain, specified as a real vector. The number of elements in the vector must be in the range [1, size(x,1)]. The number of elements in the vector must also be greater than OverlapLength.

Data Types: single | double

Number of samples overlapped between adjacent windows, specified as an integer in the range [0, size(Window,1)).

Data Types: single | double

Number of bins used to calculate the DFT of windowed input samples, specified as a positive scalar integer. If unspecified, FFTLength defaults to the number of elements in the Window.

Data Types: single | double

Frequency range in Hz, specified as a two-element row vector of increasing real values in the range [0, f/2].

Data Types: single | double

Spectrum type, specified as "power" or "magnitude":

  • "power" –– The spectral flatness is calculated for the one-sided power spectrum.

  • "magnitude" –– The spectral flatness is calculated for the one-sided magnitude spectrum.

Data Types: char | string

Output Arguments

collapse all

Spectral flatness, returned as a scalar, vector, or matrix. Each row of flatness corresponds to the spectral flatness of a window of x. Each column of flatness corresponds to an independent channel.

Spectral arithmetic mean, returned as a scalar, vector, or matrix. Each row of arithmeticMean corresponds to the arithmetic mean of the spectrum of a window of x. Each column of arithmeticMean corresponds to an independent channel.

Spectral geometric mean, returned as a scalar, vector, or matrix. Each row of geometricMean corresponds to the geometric mean of the spectrum of a window of x. Each column of geometricMean corresponds to an independent channel.

Algorithms

The spectral flatness is calculated as described in [1]:

flatness=(k=b1b2sk)1b2b11b2b1k=b1b2sk

where

  • sk is the spectral value at bin k.

  • b1 and b2 are the band edges, in bins, over which to calculate the spectral spread.

References

[1] Johnston, J. D. "Transform Coding of Audio Signals Using Perceptual Noise Criteria." IEEE Journal on Selected Areas in Communications. Vol. 6, Number 2, 1988, pp. 314–323.

Extended Capabilities

C/C++ Code Generation
Generate C and C++ code using MATLAB® Coder™.

GPU Arrays
Accelerate code by running on a graphics processing unit (GPU) using Parallel Computing Toolbox™.

Version History

Introduced in R2019a