INSTITUTO TECNOLÓGICO Y DE ESTUDIOS SUPERIORES DE MONTERREY
Project: Noise Cancellation (Filtering and STFT)
Tema Members:
Dan Fong, Luis Arturo
Rangel García, Frida Berenice
Villicaña Ibargüengoytia, José Rubén
Zhu Chen, Alfredo
Abstract:
This repositroy contains the project form the Laboratory of Digital Signal Processing. It was developed a noise cancellation algorithm by using filters and the STFT.
The first approach consists on using notch filters to attenuate the frequency componentes of the signal where the noise is allocated.
The second approach uses the Short-time Fourier Transform (STFT) to analyse the signal in three dimentios: Time, frequency and aplitude. It is also knwon as Spectrogram. Mathematical opperation was applied to cancel unwanted noise.
There are a MATLAB script and a MATLAB live script for each approach in this folder. An Graphical User Interface (GUI) was also developed for the user. To use it, a MATLAB app should be installed.
Users can use any audio to run the program. To demostrate its functionality, six different audios were provided. This algorithm works mainly with two audios; the first one is only focused on the noise of the room (noise sensing), the second one is the message (voice, instrument, input signal, etc...) with the noise, naturally. It does not matter if the noise caputred on the second sample do not match with the noise obtained on the first audio (Actually, it needs to be different).
The samples must be with extention ".wav" and the audio's name shoud be cahnged on MATLAB's scripts.
On the app, both approches can be found (same file for filtering and STFT application).
The ".prj" for the app is also included if the user wants to modify the parameters.