Audio Pitch Shift Using Audio Capture Playback

Pitch Shifting

Pitch shifting is the ability to modify the pitch of an audio signal to either increase or decrease the pitch. For example, when a fast car passes you in the street, the pitch of the sound from the car increases as the vehicle approaches you and decreases as the vehicle moves away from you. As the source of the audio moves closer or further away from the receiver, the successive signals reach the receiver at either smaller or larger intervals of time. This time difference causes change in the frequency of the audio being heard by the receiver. You can perform a similar pitch-shifting operation by splitting an audio signal into two components, delaying the signals using varying time periods, and then adding them back together. This process simulates the effect of either increasing or decreasing the pitch depending on how the two signals are delayed or overlapped. To ensure uniform power levels, the individual gains for the split signals must be modulated.

Required Products

Required Hardware

Step 1: Connect the Raspberry Pi Hardware for Pitch Shift

Before you start this example, we recommend you to complete the Getting Started with MATLAB Support Package for Raspberry Pi Hardware example.

1. Connect the micro-USB end of the USB cable to the Raspberry Pi and the regular USB end of the USB cable to the computer. Wait until the PWR LED on the hardware starts blinking.

2. Connect the webcam or the microphone to one of the USB ports on the hardware. Note that some webcams draw too much power and may require a powered USB hub for proper operation.

3. Connect a pair of headphones to the 3.5 mm headphone jack of the hardware.

4. Using the Hardware Setup screen, configure the Raspberry Pi network.

Step 2: Change the Audio Output Mode to Headphone Jack

1. In the hardware terminal, execute this command to open the Software Configuration Tool of the hardware.

sudo raspi-config

2. In the Software Configuration Tool Window, select Advanced Options and press Enter, then select Audio and press Enter.

3. Select the Force 3.5mm ('headphone') jack option and press Enter.

4. Press OK.

Step 3: Create Capture and Playback Objects

Create a raspi object.

rpi = raspi();

Create captureObj and playerObj system objects for audio processing on Raspberry Pi Hardware.

The captureObj is a connection to the audio source device plughw:1,0. The audio input is captured at a sample rate of 48000, and samples per frame of 4800.

captureObj = audiocapture(rpi,'plughw:1,0','SampleRate', 48000, 'SamplesPerFrame', 4800);

The playbackObj is a connection to the audio player device plughw:0,1. The audio output is played at a sample rate of 48000.

playbackObj = audioplayer(rpi,'plughw:0,1', 'SampleRate', 48000);

If you do not know the name of the device, use the listAudioDevices function attached with this example.

Step 4: Perform Pitch Shift of Audio Input

Delay-Based Pitch Shifter (Audio Toolbox) is an audio plugin object that performs the pitch shifting algorithm. The plugin parameters are the pitch shift (in semi-tones), the cross-fading factor (which controls the overlap between the two delay branches), and the sampling frequency. The plugin's parameters are tuned by setting their values to the input arguments pitch and overlap, respectively.

pitch = -5;
overlap = 0.2;
Fs = 8192;

shiftPitch, a function that may be used to perform pitch shifting, instantiates an audiopluginexample.PitchShifter plugin, and uses the setSampleRate method to set its sampling rate to the input argument Fs.

pitchShifter = audiopluginexample.PitchShifter('PitchShift',8,'Overlap',0.3);
setSampleRate(pitchShifter,Fs);

Capture audio from the input device. The input is of type int16 and needs to be converted to type double before processing the data. This is because the function shiftPitch expects all its inputs to be of the same data type. The output data needs to be of type int16 and thus pitchShifted which is a double is cast to be of type int16 before being sent to audio playback device.

for k = 1:3000
    input = capture(captureObj);
    pitchShifted = zeros(size(double(input)),'like',double(input));
    pitchShifter.PitchShift = pitch;
    pitchShifter.Overlap = overlap;
    [pitchShifted] = pitchShifter(double(input));
    play(playbackObj,int16(pitchShifted));
end

Step 5: Deploy the MATLAB Function

You can deploy the raspi_pitchshiftdeployment() function on the hardware.

function raspi_pitchshiftdeployment()
  rpi = raspi();
  captureObj = audiocapture(rpi,'plughw:1,0','SampleRate', 48000, 'SamplesPerFrame', 4800);
  playbackObj = audioplayer(rpi,'plughw:0,1', 'SampleRate', 48000);

   pitch = -5;
   overlap = 0.2;
   Fs = 8192;

   pitchShifter = audiopluginexample.PitchShifter('PitchShift',8,'Overlap',0.3);
   setSampleRate(pitchShifter,Fs);

  for k = 1:3000
    input = capture(captureObj);
    pitchShifted = zeros(size(double(input)),'like',double(input)); %#ok<PREALL>
    pitchShifter.PitchShift = pitch;
    pitchShifter.Overlap = overlap;
   [pitchShifted] = pitchShifter(double(input));
   play(playbackObj,int16(pitchShifted));
  end
 end

Deploy the raspi_pitchshiftdeployment function as a standalone executable on the hardware by using the deploy function.

board = targetHardware('Raspberry Pi')
deploy(board,'raspi_pitchshiftdeployment')

Code generation successful: View report

The deploy function initiates code generation of the raspi_pitchshiftdeployment function. At the end of code generation, MATLAB generates a code generation report. Use this report to debug the raspi_pitchshiftdeployment function for any build errors and warnings in the generated code.

After successfully generating the code, the support package loads and runs the code as a standalone executable on the hardware. The executable starts acquiring live audio input from the audio device, runs the pitch shift algorithm on the acquired audio, and then plays the result on the playback device.