1. PURPOSE / DESCRIPTION
1.1 The objective of this test is to generate a white noise which could soften the sound generated by propellers and motors while playing them simultaneously.
2. TEST SET-UP
2.1 In the zip file along with this ETP, you should be able to find a folder called “Data” and a .m file named “audio_processing”.
2.2 The .m file needs to be run on a machine installed with MATLAB.
3.1. Open the file “audio_process.m” and run the MATLAB code.
3.2. The main code contains the following sections:
3.2.1. Load the correct audio file. You should input a correct path to the wavread function. Note that only wav file is supported to be read by MATLAB. The return variable Fs is the sampling rate.
3.2.2. Generate white noise. The pure white noise is generated by wgn function and normalized afterwards. It is stored in the variable w.
3.2.3. Apply high pass filter. A high pass filter is written to process the generated white noised. You can modify the cut off frequency by changing the value of variable Fh.
3.2.4. Apply low pass filter. A low pass filter is written to process the generated white noised. You can modify the cut off frequency by changing the value of variable Fl.
3.2.5. Apply notch filters. A few notch filters is written to process the generated white noised. You can add target frequencies in the array notch. For each frequency in the array, the program will generate a notch filter to process the audio. Note that the ordering of the frequencies in the array may influence the performance.
3.2.6. Play the sound. After processing, the sound could be played by using sound function. Note that you can play two sounds simultaneously by two successive calls.
3.3. According the requirements, a few figures are generated to show the results.
In the picture above, the top two figures show the original sound in time domain and frequency domain, respectively. The bottom two figures show the modified white noise in time domain and frequency domain, respectively.
4. ACCEPTANCE CRITERIA
4.1 You can play the generated sound to evaluate the final results. The new white noise should be able to soften the original humming sound.