Chetan

To Design A Filter To Remove The Noise From Audio Signals.

Project Submitted by Chetan Reddy Gaddam

11/30/09

Analog And Digital Filter Design

ENEE 4554

Abstract

The Main Objective Of the Project is to reduce the unwanted background noise,particularly in Audio music Signals, which is significantly reduced by a filtering system, which is operated in response to the fundamental and harmonic content and degree of correlation or non randomness of the audio signal. A voltage controlled analog filters are established in the audio signal path for individually and selectively passing frequency content of the audio signal falling within each frequency band. When the signal is passed through the filter, detected by band pass signal detectors, the output of which are connected to and for operating associated band pass gates. This gates are closed blocking noise present on incoming signal. These detectors sense the presence of fundamental signal and causes the appropriate gate to open and pass the signal.

Introduction

In General the audio signal comprises of background noise and music signal. The presence of background noise accompanying audio signals is undesirable but is also unavoidable by-product of transmission of audio signals. Filter is a device which removes from a signal some unwanted component or feature. Filtering is a class of signal processing, the defining feature of filters being the complete or partial suppression of some aspect of the signal. Most often, this means removing some frequencies and not others in order to suppress interfering signals and reduce background noise. However, filters do not exclusively act in the frequency domain especially in the field of image processing many other targets for filtering exist. There are various filtering techniques used to reduce the noise content, which by increasing the amplitude of the signal. Dynamic filtering is a technique mostly used for reducing the noise content in an audio signal, the amount of roll off or cutoff of the high frequencies is adjusted by a control voltage which is a function of energy content of the transmitted musical signal. These dynamic filters represent an improvement over the simple traditional filter, however, known dynamic filtering techniques only remove the higher frequency noise and cause a certain amount of undesirable modulation of the audible higher frequencies.

Data Analysis There are six audio music files given and they are corrupted by some noise. Filtering Of Marimba-Noisy.wav is explained in detail. Marimba- noisy.wav is an audio file comprising of musical signal along with background noise.

The above graph consists of 1.Magnitude plot : this graph is plotted against frequency Vs magnitude.

2. Plot of input signal: this graph is plotted for the given input signal. 3.Spectrogram for the given input signal: A spectrogram is an image that shows how the spectral density of a signal varies with time. spectrograms are used to identify phonetic sounds, to analyze the cries of animals, and in the fields of music, sonar/radar, speech processing, seismology, etc. The instrument that generates a spectrogram is called a spectrograph or sonograph. A graph is obtained against time Vs frequency 4.PSD for the given input signal: PSD is a positive real function of a frequency variable associated with a stationary stochastic process, or a deterministic function of time, which has dimensions of power per Hz, or energy per Hz. A graph is plotted against frequency Vs power spectra magnitude.

This is the Filtered output .

Sampling rate : 48000 Data size:

Types of filters 1.Bessel filter

220500

2.Butterworth filter 3.chebyshev filter(I&II) 4.Elliptical filter Elliptical filter: An elliptic filter is a signal processing filter with equalized ripple behavior in both the passband and the stopband. The amount of ripple in each band is independently adjustable, and no other filter of equal order can have a faster transition in gain between the passband and the stopband, for the given values of ripple. Alternatively, one may give up the ability to independently adjust the passband and stopband ripple, and instead design a filter which is maximally insensitive to component variations. As the ripple in the stopband approaches zero, the filter becomes a type I Chebyshev filter. As the ripple in the passband approaches zero, the filter becomes a type II Chebyshev filter and finally, as both ripple values approach zero, the filter becomes a Butterworth filter.

Filter Design Specifications: 1.Order(n): The order of the filter is chosen to be 8. 2.Pass Band Ripple(Rp): pass band ripples are the ripples obtained in pass band region. Rp is chosen to be 1. 3.stop band ripple(Rs): stop band ripples are the ripples obtained in stop band. Rs is chosen to be 50 4.cut off frequency: cut off frequency is a value after which the signal is attenuated. It is chosen to be 1000/(r/2),where r is the sampling rate of the given audio signal.

Filter Design & Performance Results The given signal is read using the command [d,r]=wavread(‘input signal’)

Which gives the sampling rate and data size, and using them input signal,spectrogram & Power spectral density are plotted. By using the command soundsc(d,r),the audio sound is obtained and by using the desired filter specifications, a filter is designed to filter the noise in the input signal, here a low pass elliptical filter is used,by which a better ouput signal with less noise is obtained compared with other filters. The graphs are plotted for the filtered signal and the sound of the output filtered signal is saved using a wavwrite function. Performance result: Input signal Giga-noisy.wav

Filtered signal

Sampling rate: 44100 Data size: 220500

Conclusion:

The given audio signal is analysed and filtered using elliptical filter and the results showed that the audio signal has reduced noise.From the above graphs we can differenciate between given audio signal and the filtered signal.

Reference: 1.Digital Signal Processing by Ramesh babu. 2. Analog and digital filter design by Steve Winder.