crossoverFilter
Audio crossover filter
Description
The crossoverFilter
System object™ implements an audio crossover filter, which is used to split an audio signal
into two or more frequency bands. Crossover filters are multiband filters whose overall
magnitude frequency response is flat.
To implement an audio crossover filter:
Create the
crossoverFilterobject and set its properties.Call the object with arguments, as if it were a function.
To learn more about how System objects work, see What Are System Objects?
Creation
Syntax
Description
crossFilt = crossoverFilter creates a System object, crossFilt, that implements an audio crossover
filter.
crossFilt = crossoverFilter(
sets the NumCrossovers
property to nCrossovers)nCrossovers.
crossFilt = crossoverFilter(
sets the CrossoverFrequencies property to nCrossovers,xFrequencies)xFrequencies.
crossFilt = crossoverFilter(
sets the CrossoverSlopes property to nCrossovers,xFrequencies,xSlopes)xSlopes.
crossFilt = crossoverFilter(
sets the SampleRate
property to nCrossovers,xFrequencies,xSlopes,Fs)Fs.
crossFilt = crossoverFilter(___,
sets each property Name,Value)Name to the specified Value.
Unspecified properties have default values.
Example: crossFilt =
crossoverFilter(2,'CrossoverFrequencies',[100,800],'CrossoverSlopes',[6,48])
creates a System object, crossFilt, with two crossovers located at 100 Hz and 800
Hz, and crossover slopes of 6 dB/octave and 48 dB/octave, respectively.
Properties
Usage
Description
[band1,...,bandN] =
crossFilt(audioIn)audioIn, and returns the
filtered output bands, [band1,...,bandN], where N =
.NumCrossovers + 1
Input Arguments
Output Arguments
Object Functions
To use an object function, specify the
System object as the first input argument. For
example, to release system resources of a System object named obj, use
this syntax:
release(obj)
The createAudioPluginClass and configureMIDI
functions map tunable properties of the crossoverFilter
System object to user-facing parameters:
| Property | Range | Mapping | Unit |
|---|---|---|---|
CrossoverFrequencies | [20, 20000] | linear | Hz |
CrossoverSlopes | [6, 48] | linear | dB/octave |
Examples
Use the crossoverFilter object to split Gaussian noise into three separate frequency bands.
Create a 5 second noise signal that assumes a 24 kHz sample rate.
fs = 24e3; noise = randn(fs*5,1);
Create a crossoverFilter object with 2 crossovers (3 bands), crossover frequencies at 4 kHz and 8 kHz, a slope of 48 dB/octave, and a sample rate of 24 kHz.
crossFilt = crossoverFilter( ... NumCrossovers=2, ... CrossoverFrequencies=[4000,8000], ... CrossoverSlopes=48, ... SampleRate=fs);
Visualize the magnitude response of your crossover filter object.
visualize(crossFilt)
Call your crossover filter like a function with the noise signal as the argument.
[y1,y2,y3] = crossFilt(noise);
Visualize the results using a spectrogram.
tiledlayout(4,1) nexttile stft(noise,fs,FrequencyRange="onesided") title("Noise") nexttile stft(y1,fs,FrequencyRange="onesided") title("y1") nexttile stft(y2,fs,FrequencyRange="onesided") title("y2") nexttile stft(y3,fs,FrequencyRange="onesided") title("y3")
Use the crossoverFilter object to split an audio signal into three frequency bands.
Create the dsp.AudioFileReader and audioDeviceWriter objects. Use the sample rate of the reader as the sample rate of the writer.
samplesPerFrame = 256; fileReader = dsp.AudioFileReader( ... "RockGuitar-16-44p1-stereo-72secs.wav", ... SamplesPerFrame=samplesPerFrame); deviceWriter = audioDeviceWriter( ... SampleRate=fileReader.SampleRate);
Create a crossoverFilter object with 2 crossovers (3 bands), crossover frequencies at 500 Hz and 1 kHz, and a slope of 18 dB/octave. Use the sample rate of the reader as the sample rate of the crossover filter.
crossFilt = crossoverFilter( ... NumCrossovers=2, ... CrossoverFrequencies=[500,1000], ... CrossoverSlopes=18, ... SampleRate=fileReader.SampleRate);
Visualize the bands of the crossover filter.
visualize(crossFilt)
Get the cost of the crossover filter.
cost(crossFilt)
ans = struct with fields:
NumCoefficients: 48
NumStates: 18
MultiplicationsPerInputSample: 48
AdditionsPerInputSample: 37
Create a spectrum analyzer to visualize the effect of the crossover filter.
scope = spectrumAnalyzer( ... SampleRate=fileReader.SampleRate, ... PlotAsTwoSidedSpectrum=false, ... FrequencyScale="log", ... Title="Crossover Bands and Reconstructed Signal", ... ShowLegend=true, ... ChannelNames=["Original Signal","Band 1","Band 2","Band 3","Sum"]);
Play 10 seconds of the audio signal. Visualize the spectrum of the original audio, the crossover bands, and the reconstructed signal (sum of bands).
setup(scope,ones(samplesPerFrame,5)) count = 0; while count < (fileReader.SampleRate/samplesPerFrame)*10 originalSignal = fileReader(); [band1,band2,band3] = crossFilt(originalSignal); sumOfBands = band1 + band2 + band3; scope([originalSignal(:,1), ... band1(:,1), ... band2(:,1), ... band3(:,1), ... sumOfBands(:,1)]) deviceWriter(sumOfBands); count = count + 1; end release(fileReader) release(crossFilt) release(deviceWriter) release(scope)
De-essing is the process of diminishing sibilant sounds in an audio signal. Sibilance refers to the s, z, and sh sounds in speech, which can be disproportionately emphasized during recording. es sounds fall under the category of unvoiced speech with all consonants and have a higher frequency than voiced speech. In this example, you apply split-band de-essing to a speech signal by separating the signal into high and low frequencies, applying an expander to diminish the sibilant frequencies, and then remixing the channels.
Create a dsp.AudioFileReader object and an audioDeviceWriter object to read from a sound file and write to an audio device. Listen to the unprocessed signal. Then release the file reader and device writer.
fileReader = dsp.AudioFileReader( ... 'Sibilance.wav'); deviceWriter = audioDeviceWriter; while ~isDone(fileReader) audioIn = fileReader(); deviceWriter(audioIn); end release(deviceWriter) release(fileReader)
Create an expander System object to de-ess the audio signal. Set the sample rate of the expander to the sample rate of the audio file. Create a two-band crossover filter with a crossover of 3000 Hz. Sibilance is usually found in this range. Set the crossover slope to 12. Plot the frequency response of the crossover filter to confirm your design visually.
dRExpander = expander( ... 'Threshold',-50, ... 'AttackTime',0.05, ... 'ReleaseTime',0.05, ... 'HoldTime',0.005, ... 'SampleRate',fileReader.SampleRate); crossFilt = crossoverFilter( ... 'NumCrossovers',1, ... 'CrossoverFrequencies',3000, ... 'CrossoverSlopes',12); visualize(crossFilt)
Create a timescope object to visualize the original and processed audio signals.
scope = timescope( ... 'SampleRate',fileReader.SampleRate, ... 'TimeSpanOverrunAction','Scroll', ... 'TimeSpanSource','Property','TimeSpan',4, ... 'BufferLength',fileReader.SampleRate*8, ... 'YLimits',[-1 1], ... 'ShowGrid',true, ... 'ShowLegend',true, ... 'ChannelNames',{'Original','Processed'});
In an audio stream loop:
Read in a frame of the audio file.
Split the audio signal into two bands.
Apply dynamic range expansion to the upper band.
Remix the channels.
Write the processed audio signal to your audio device for listening.
Visualize the processed and unprocessed signals on a time scope.
As a best practice, release your objects once done.
while ~isDone(fileReader) audioIn = fileReader(); [band1,band2] = crossFilt(audioIn); band2processed = dRExpander(band2); procAudio = band1 + band2processed; deviceWriter(procAudio); scope([audioIn procAudio]); end release(deviceWriter) release(fileReader) release(scope)
release(crossFilt) release(dRExpander)
Plosives are consonant sounds resulting from a sudden release of airflow. They are most pronounced in words beginning with p, d, and g sounds. Plosives can be emphasized by the recording process and are often displeasurable to hear. In this example, you minimize the plosives of a speech signal by applying highpass filtering and low-band compression.
Create a dsp.AudioFileReader object and an audioDeviceWriter object to read an audio signal from a file and write an audio signal to a device. Play the unprocessed signal. Then release the file reader and device writer.
fileReader = dsp.AudioFileReader('audioPlosives.wav'); deviceWriter = audioDeviceWriter('SampleRate',fileReader.SampleRate); while ~isDone(fileReader) audioIn = fileReader(); deviceWriter(audioIn); end release(deviceWriter) release(fileReader)
Design a highpass filter with a steep rolloff of all frequencies below 120 Hz. Use a dsp.SOSFilter object to implement the highpass filter design. Create a crossover filter with one crossover at 250 Hz. The crossover filter enables you to separate the band of interest for processing. Create a dynamic range compressor to compress the dynamic range of plosive sounds. To apply no make-up gain, set the MakeUpGainMode to "Property" and use the default 0 dB MakeUpGain property value. Create a time scope to visualize the processed and unprocessed audio signal.
[B,A] = designVarSlopeFilter(48,120/(fileReader.SampleRate/2),"hi",Orientation="row"); biquadFilter = dsp.SOSFilter(B,A); crossFilt = crossoverFilter( ... "SampleRate",fileReader.SampleRate, ... "NumCrossovers",1, ... "CrossoverFrequencies",250, ... "CrossoverSlopes",48); dRCompressor = compressor( ... "Threshold",-35, ... "Ratio",10, ... "KneeWidth",20, ... "AttackTime",1e-4, ... "ReleaseTime",3e-1, ... "MakeUpGainMode","Property", ... "SampleRate",fileReader.SampleRate); scope = timescope( ... "SampleRate",fileReader.SampleRate, ... "TimeSpanSource","property","TimeSpan",3, ... "BufferLength",fileReader.SampleRate*3*2, ... "YLimits",[-1 1], ... "ShowGrid",true, ... "ShowLegend",true, ... "ChannelNames",{'Original','Processed'});
In an audio stream loop:
Read in a frame of the audio file.
Apply highpass filtering using your biquad filter.
Split the audio signal into two bands.
Apply dynamic range compression to the lower band.
Remix the channels.
Write the processed audio signal to your audio device for listening.
Visualize the processed and unprocessed signals on a time scope.
As a best practice, release your objects once done.
while ~isDone(fileReader) audioIn = fileReader(); audioIn = biquadFilter(audioIn); [band1,band2] = crossFilt(audioIn); band1compressed = dRCompressor(band1); audioOut = band1compressed + band2; deviceWriter(audioOut); scope([audioIn audioOut]) end
As a best practice, release your objects once done.
release(deviceWriter) release(fileReader) release(crossFilt) release(dRCompressor) release(scope)
Create a dsp.AudioFileReader to read in audio frame-by-frame. Create a audioDeviceWriter to write audio to your sound card. Create a crossoverFilter to process the audio data. Call visualize to plot the frequency responses of the filters.
frameLength = 1024; fileReader = dsp.AudioFileReader('RockDrums-44p1-stereo-11secs.mp3', ... 'SamplesPerFrame',frameLength); deviceWriter = audioDeviceWriter('SampleRate',fileReader.SampleRate); xFilt = crossoverFilter('SampleRate',fileReader.SampleRate); visualize(xFilt)
Call parameterTuner to open a UI to tune parameters of the crossover filter while streaming.
parameterTuner(xFilt)
In an audio stream loop:
Read in a frame of audio from the file.
Apply crossover filtering.
Write the frame of audio to your audio device for listening.
While streaming, tune parameters of the crossover filter and listen to the effect.
while ~isDone(fileReader) audioIn = fileReader(); [low,high] = xFilt(audioIn); deviceWriter([low(:,1),high(:,1)]); drawnow limitrate % required to update parameter end
As a best practice, release your objects once done.
release(deviceWriter) release(fileReader) release(xFilt)
Algorithms
The crossover System object is implemented as a binary tree of crossover pairs with additional phase-compensating sections [1]. Odd-order crossovers are implemented with Butterworth filters, while even-order crossovers are implemented with cascaded Butterworth filters (Linkwitz-Riley filters).
Odd-order two-band (one crossover) filters are implemented as parallel complementary highpass and lowpass filters.
LP and HP are Butterworth filters of order N, implemented as direct-form Ⅱ transposed second-order sections. The shared cutoff frequency used in their design corresponds to the crossover of the resulting bands.
Even-order two-band (one crossover) filters are implemented as parallel complementary highpass and lowpass filters.
LP and HP are Butterworth filters of order N/2, where N is the order of the overall filter. The filters are implemented as direct-form Ⅱ transposed second-order sections.
For overall filters of orders 2 and 6, XHI is multiplied by –1 internally so that the branches of your crossover pair are in-phase.
Even-order three-band (two crossovers) filters are implemented as parallel complementary highpass and lowpass filters organized in a tree structure.
The phase-compensating section is equivalent to an allpass filter.
The design of four-band and five-band filters (three and four crossovers) are extensions of the pattern developed for even-order and odd-order crossovers and the tree structure specified for three-band (two crossover) filters.
References
[1] D’Appolito, Joseph A. "Active Realization of Multiway All-Pass Crossover Systems." Journal of Audio Engineering Society. Vol. 35, Issue 4, 1987, pp. 239–245.
